Quantitative determination of the myofibrillar proteins and connective

pounds in typical porcine skeletal muscle tissues, in pu- rified actin, and the two major intracellular muscle protein and connective tissue fractions...
0 downloads 0 Views 3MB Size
J. Agric. FoodChem. 1988, 36, 1131-1146

Proteins“. Can. J. Biochem. 1975,53, 96-101. Zarkadas, C. G. “Accelerated Methods for Determining Methylated Basic Amino Acid in Proteins and Muscle Tissues and Stable Crosslinks in Elastin and Collagen”. J. Appl. Biochem. 1979,I , 148-163. Zarkadas, C. G. “The Determination of the Myofibrillar and Connective Tissue Contents of Meats and Composite Meat Products”; Proceedings of the 27th European Fleischforscher-Kongress IT, Vienna, Austria, 1981;pp 572-575. Zarkadas, C. G.; Hulan, H. W.; Proudfoot, F. G. “A Comparison of the Amino Acid Composition of Two Commercial Oat Groates”. Cereal Chem. 1982, 159, 323-327. Zarkadas, C. G.; Zarkadas, G. C.; Karatzas, C. H.; Khalili, A. D.; Nguyen, Q. “Rapid Method for Determining Desmosine, Isodesmosine, 5-Hydroxylysine,Triptophan, Lysinoalanine and the Amino-sugars in Proteins and Tissues”. J . Chromatogr. 1986,378,67-76. Zarkadas, C. G.; Marshall, W. D.; Khalili, A. D.; Nguyen, Q.; Zarkadas, G. C.; Karatzas, C. N.; Khanizadeh, S. “Mineral Composition of Selected Bovine, Porcine and Avian Muscles,

1131

and Meat Products”. J. Food Sci. 1987a,52,520-525. Zarkadas, C. G.; Rochemont, J. A,; Zarkadas, G. C.; Karatzas, C. N.; Khalili, A. D. “Determination of Methylated Basic, 5Hydroxylsine, Elastin Crosslins, Other Amino Acids and the Amino Sugars in Proteins and Tissues”. Anal. Biochem. 1987b, 160, 251-266.

Zarkadas, C. G.; Meighen, E. A,; Zarkadas, G. C.; Karatzas, C. N.; Rochemont,J. A.; Berthelet, M. “The Determination of the Myofibrillar and Connective Tissue Proteins in the Bovine Diaphragm”. J. Agric. Food Chem. 1988,companion paper in this issue. Received for review January 27,1987. Accepted November 24, 1987. This investigation was supported by Agriculture Canada (DSS Contract No. DSU82-00523with McGill University) and was carried out at the former Department of Agricultural Chemistry and Physics, Macdonald College of McGill University; and by McGill University for a graduate Bursary awarded to C.N.K. (1984-1985).

Quantitative Determination of the Myofibrillar Proteins and Connective Tissue Content in Selected Porcine Skeletal Muscles’ Constantinos G. Zarkadas,* Constantinos N. Karatzas, Ali D. Khalili, Shahrokh Khanizadeh, and Guylaine Morin

The new analytical chromatographic methods developed to quantitate the unique amino acids that occur in proteins have been successfully applied for the determination of the myofibrillar and connective tissue content of both select porcine skeletal muscles and the intracellular and extracellulw protein fractions. The proposed chemical approach is based on the direct determination of the myofibrillar myosin and actin contents of skeletal muscles from the amounts of protein bound NT-methylhistidine present. Collagen and collagen-like proteins can be calculated from the amounts of 5-hydroxylysine found and the elastin content from the amounts of desmosine or isodesmosine present. These quantitations are based on the total protein content of the selected porcine muscles determined by their detailed amino acid composition. Actin accounts for an estimated 10.2-11.5% of the total porcine muscle mass or about 21.1% of the total myofibrillar protein, while myosin ranged from 21.3 to 24.0% of the total muscle protein corresponding to 43.9% of the myofibrillar proteins (52.14% of protein). Total porcine muscle collagen ranged from 2.84 to 5.89% in select porcine muscles, while elastin accounts for an estimated 0.063-0.143%.

The possibility of using NT-methylhistidine[His(.r-Me)] as an index for determining the absolute mass of the myofibrillar proteins myosin and actin in skeletal muscles and composite meat products has stimulated considerable interest recently (Hibbert and Lawrie, 1972; Olsman and Slump, 1981; Ranken, 1984; Expert Work Group, FSIS, 1984; Benedict, 1987; Ashworth, 1987; McNeal, 1987). This quantitation is based on the following findings. Sequence studies have shown that actin contains 1mol of His(.r-Me) at position 73 in its amino sequence (Elzinga et al., 1973; Vanderkerckhoveand Weber, 1978,1979)and that myosin isolated from adult fast-twitch white skeletal muscles contains 1mol of His(.r-Me) at position 755 in each of the two heavy chains of this protein (Okamoto and Yount, Agriculture Canada’s Muscle Biochemistry Laboratory, Saint Hyacinthe Food Research Centre, Research Branch, Eastern Region, and The McGill Nutrition and Food Science Centre, Macdonald College of McGill University, Ste. Anne de Bellevue, Quebec, Canada H9X 1CO. Contribution No. 29, St. Hyacinthe Food Research Centre. 0021-8561/88/1436-1131$01.50/0

1975; Maita et al., 1987). Other studies have indicated that His(.r-Me) is absent from all other muscle and nonmuscle proteins (Huszar, 1984). In addition, Yates and Greaser (1983) have shown that the psoas skeletal muscle myosin accounts for 43% and actin for an estimated 22% of the myofibrillar protein mass (57.7%) of skeletal muscle. Moreover, the in situ molar ratio of actin to myosin in the myofibrils of skeletal muscle has been shown to be 6/1 (Murakami and Uchida, 1985). Measurements of the levels of His(.r-Me) in several experimental animals (Asatoor and Armstrong, 1967; Johnson et al., 1967; Haverberg et al., 1975; Holbrook et al., 1979) and in bovine, ovine, and avian skeletal muscle tissues (Rangeley and Lawrie, 1976; Olsman and Slump, 1981; White and Lawrie, 1985; Jones et al., 1985, 1987), using a variety of chromatographicmethods (Hancock and Harding, 1984; Ashworth, 1987), have yielded variable amounts of His(.r-Me) among the muscle tissues studied, ranging from 2.24 to 10.6 pmol of His(.r-Me)/g of tissue. Although some of this variation was attributed to the distribution of porcine muscle fiber types or the presence of variable amounts of balenine (Carnegie et al., 1982,1984; Harris and Milne, 1981, 1987), a histidine dipeptide, p-

Published 1988 by the American Chemical Society

1132 J. Agric. Food Chem., Vol. 36, No. 6, 1988

alanyl-L-W-methylhistine, which must be extracted from muscle tissues prior to acid hydrolysis, most of this variation appears to be due to incomplete separation of such small amounts of His(7-Me) from other components present in such complex muscle tissue hydrolysate by the multicolumn systems employed (Hancock and Harding, 1984). This paper describes the application of the new analytical chromatographic methods developed (Zarkadas et al., 1986,1987b)to quantitate the unique methylated basic amino acids, including 5-hydroxylysine [Lys(5-OH)], desmosine [Des], isodesmosine [iDes], and related compounds in typical porcine skeletal muscle tissues, in purified actin, and the two major intracellular muscle protein and connective tissue fractions, prepared and quantitated by the methods of McCollester (1962) and Laurent et al. (1981). The aim was to determine whether the levels of these unique basic amino acids in the selected porcine skeletal muscle tissues could be used for the determination of their myofibrillar and connective tissue proteins from the amounts of protein-bound His(r-Me) and Lys(5-OH) found in their acid hydrolysates, respectively. The elastin content in muscles can also be calculated from the amounts of Des or iDes found (Zarkadas, 1981; Zarkadas et al., 1986). These calculations are based on the total protein content of porcine muscle tissues and the isolated intracellular muscle protein and extracellular matrix fractions determined by their detailed amino acid compositions. MATERIALS AND METHODS

Materials. Types DC-4A (Lot No. 750) and DC-5A (Lot No. 746) cation-exchange spherical resins, sized to 9.0 f 0.5 and 6.0 f 0.5 pm, respectively, were purchased from Dionex Chemical Co., Sunnyvale, CA. Type AA-10 resin, which is equivalent to Dionex Type DC-4A resin, and Type I standard amino acid calibration mixture were obtained from Beckman Instruments, Inc., Palo Alto, CA. The diastereoisomer mixture of 5-hydroxy-D~-lysine,N6methyl-L-lysine, N6,NG-dimethyl-L- and N6,P,IP-trimethyl-L-lysines bis@-hydroxyazobenzenesulfonate) monohydrate, D-glucosamine monohydrochloride, Dgalactosamine monohydrochloride, N'-methyl-~-histidine, and N"-methyl-L-histidine monohydrate were purchased from Calbiochem-Behring Corp., La Jolla, CA. DLOrnithine was purchased from Schwarg/Mann, Orangeburg, NY. Norleucine and L-2-amino-3-guanidinopropionic acid was from Pierce Chemical Co., Rockford, IL, and 3-nitro-~-tyrosine from Aldrich Chemical Co., Milwaukee, WI. Because Des and iDes are not commercially available, these cross-linking amino acids were isolated by the preparative method described by Zarkadas (1979) using bovine Ligamentum nuchae elastin purchased from Sigma Chemical Co., St. Louis, MO. Sodium dodecyl sulfate (SDS) was also from Sigma. The highly purified citrate buffers (pH 3.28,0.20 M; pH 4.25,0.20 M; pH 6.40,l.O M), sample dilution buffer (pH 2.2, 0.20 M), and ninhydrin recommended for high-sensitivity standard single-microcolumn analysis were purchased as either concentrates or ready-to-use solutions from Beckman Instruments. The chemicals and reagents employed for determining the methylated basic amino acids were purchased from Pierce. Sephadex G-10, G-25, and G-200 (fine bead type) were obtained from Pharmacia Fine Chemicals, Uppsala, Sweden. AG-500-X8(D) analytical-grade mixed bed (20-50 mesh fully regenerated), acrylamide (>99.9% purity), N,N'-methylenebisacrylamide, TEMED (N,N,N',N'tetramethylethylenediamine), ammonium persulfate, Coomassie Brilliant Blue G-250 and R-250, Bromophenol Blue, glycine, Bio-Lyte (pH 3-10), and sodium dodecyl

Zarkadas et al.

sulfate (SDS) were purchased from Bio-Rad Laboratories, Richmond, CA. Ampholine, pH 3.5-10 (LKB, Broma, Sweden), was obtained from Fisher Scientific Canada, Montreal, Quebec. Electrophoresis calibration Kit H (Lot No. 9015) for molecular weight determination was obtained from Pharmacia. The 2-mercaptoethanol was obtained from Eastman Kodak Co., Rochester, NY. Nonidet P40 was supplied from BDH Chemicals Ltd., Poole, England. High-purity urea were purchased from SchwarzIMann, which was further purified by passing through an AG 501-XS(D) mixed-bed column (5 X 20 cm). The distilled water used in the preparation of all buffers and reagents was purified as described previously (Zarkadas et al., 1987b) and stored in glass until use. All other chemicals and reagents were of the highest purity commercially available and were used without further purification. Sampling and Preparation of Muscle Tissues. The following three typical porcine skeletal muscles were selected for this investigation: semimembranosus, semitendinosus, and external sternomandibularis. These muscle tissues were excised as 10-cm-thick muscle sections from the left side of randomly selected commercial-grade porcine carcasses (approximately 63-105 kg) from female (sows) mature Yorkshire pigs supplied by North Packers Ltd., Montreal, Quebec, and Abattoir Regional Valleyfield, Inc., Valleyfield,Quebec. All muscle tissues (approximately 200 g each) were cleaned of adhering fat, cut into small cubes, ground, frozen (-175 "C), and lyophilized. The freeze-dried samples were then pulverized in a standard electrically driven end runner mill, passed through a No. 40 mesh sieve, and then stored at -20 OC in polypropylene bottles until use. Proximate and elemental analyses were carried out in each of these samples, and the results have been reported elsewhere (Zarkadas et al., 1987a). Porcine skeletal muscle tissues are known to contain variable amounts of the histidine dipeptides carnosine, anserine, and balenine (Rangeley and Lawrie, 1977; Carnegie et al., 1982, 1984; Harris and Milne, 1987). To effectively remove all these histidine dipeptides from porcine tissue, the muscle samples were extracted with a mixture of 75% ethyl alcohol in 0.1 M HC1 (Rangeley and Lawrie, 1976) as described previously (Karatzas and Zarkadas, 1988). Extraction and Quantitation of Porcine Skeletal Muscle and Connective Tissue Protein Fractions (F1 and F2). The procedure employed for the extraction and preparation of the porcine intramuscular skeletal muscle protein (Fl) and extracellular matrix connective tissue protein fractions (F2) combines the original method of McCollester (1962) and the procedures described by Laurent et al. (1981) and Light and Champion (1984) as follows. Approximately 200 g of porcine skeletal muscle samples (external sternomandibularis, semimembranosus, semitendinosus), which were excised fresh from mature female Yorkshore sows, were cut into small cubes and homogenized in 100 mL of 0.05 M CaCl, solution in a VirTis homogenizer (Model 45) for 10 s (speed set a t 30/100) at 4 OC. The homogenate was centrifuged at 4000g (SS-34, Sorvall rotor; 6000 rpm) for 20 min at 4 "C; the supernatant was decanted through eight layers of cheesecloth to trap fat particles and was retained so that muscle proteins could be removed by precipitation with 5% trichloroacetic acid and quantitated by amino acid analysis. The pellet was resuspended in 70 mL of phosphate-buffered saline (PBS; 0.15 M NaC1/0.02 M sodium phosphate buffer, pH 7.4) at 20 "C and rehomogenized for 3 min. The homogenate was again centrifuged at 4000g for 20 min, and the supernatants were combined. The

Quantttation of Contractile Protein in Porcine Muscle

pellet was resuspended in 70 mL of 2% sodium dodecyl sulfate (SDS) and rehomogenized for 3.0 min, and the supernatants were combined. The wash procedure with SDS was repeated a further four times as described for lung tissues by Laurent et al. (1981), and the separated muscle protein fractions (F1 and F2), after desalting, drying under vacuum, and powdering, were designated SDS-solubilized intracellular protein (Fl) and SDS-insoluble extracellular matrix (F2) fractions as described previously (Zarkadas et al., 1988). Actin Purification. Porcine skeletal muscle G-actin was purified from acetone-extracted semimembranosus Yorkshire sows muscle tissues by the low-salt buffer extraction method of Spudich and Watt (1971), followed by three polymerization and depolymerization steps as recommended by Pardee and Spudich (1982). The porcine semimembranosus actin was further purified by gel permeation chromatographyon a 2.6 X 95 cm Sephacryl S-200 column eluted with the depolymerizationbuffer of Pardee and Spudich (1982). The flow rate was maintained at 60 mL/h, the effluent was collected in 3.0-pL fractions, and the fractionation was monitored by absorbancy measurements at 280 nm. Protein concentration in each fraction was determined either by the method of Lowry et al. (1951) as modified by Peterson (1983) using bovine serum albumin as a standard or by the method described by Horstmann (1979). The fractions containing the G-actin from the Sephacryl S-200 column were concentrated by the polymerization method of Pardee and Spudich (1982). Prior to use, the concentrated G-actin solutions were centrifuged for 1h at lOOOOOg and the supernatants were passed through a Sephadex G-25 column (40 cm X 2.5 cm) preequilibrated with the same depolymerization buffer of Pardee and Spudich (1982) so that excess ATP and 2mercaptoethanol and other impurities could be removed. To rule out any possible effects of aging of the samples, freshly prepared and chromatographicallypurified G-actin was used. One- and Two-Dimensional Polyacrylamide Gel Electrophoresis (PAGE). One-dimensionalSDS-PAGE was performed in a Bio-Rad Model 172A analytical-electrophoresis apparatus (Bio-Rad Laboratories) equipped to accommodate 12 gel tubes according to the procedure of Weber and Osborn (1975). The technique described by Laemmli (1970) using 10% acrylamide gel with a 3% stacking gel followed by two-dimensional SDS-PAGE according to O'Farrell(l975) was applied to further test the purity of porcine semimembranosus G-actin. The Ampholine or Bio-Lyte used for isoelectric focusing in the first dimension was pH 3.5-10.0. Second-dimensionseparation was on a 5-2090 gradient gel with a 3% stacking gel. Procedures for Amino Acid Analyses. Amino acid analyses were carried out on either a conventional (Beckman Model 1 2 0 0 or an updated and fully automated amino acid analyzer (equivalent to Beckman Model 121MB). The automated instrument was equipped with a Varian Vista 402 chromatographic data reduction system (Varian Instruments Group, Walnut Creek, CA) as described previously (Zarkadas et al., 1986, 1987b). Porcine muscle tissue samples (0.1 g) were hydrolyzed in Pyrex test tubes (18 X 150 nm) under vacuum (below 10 pmHg) with 10 mL of triple-glass-distilled constantboiling HC1 (6.0 M) at 110 "C in duplicate for each of four times, 24, 48, 72, and 96 h, respectively, with the usual precautions described previously (Hunt, 1985; Nguyen et al., 1986). The data reported for serine, threonine, and tyrosine represent the average of values extrapolated to zero time of hydrolysis. The values for valine, isoleucine,

J. Agric. FoodChem., Vol. 36, No. 6, 1988

1133

leucine, and phenylalanine are average values from 48,72, and 96 h of hydrolysis. All other values are reported as the average values from 24,48,72, and 96 h of hydrolysis (Zarkadas, 1979,1981). 4-Hydroxyprolinewas determined separately from a concentrated hydrolysate (equivalent to 0.1 mg of protein/analysis) as described previously (Zarkadas et al., 1986), and recoveries were calculated relative to alanine. Methionine and cyst(e)ine were determined separately (0.1 g) by the performic acid procedure of Moore (1963) as described previously (Nguyen et al., 1986). Recovery of cyst(e)ine as cysteic acid and methionine as the dioxide was calculated in proportion to the yields obtained by the performic acid treatment of standard solutions of these amino acids and relative to alanine and leucine present in the sample. Tryptophan in porcine muscle samples (0.1 g) was determined separately after alkaline hydrolysis (Hugli and Moore, 1972) by an improved chromatographic procedure (Zarkadas et al., 1986) using 3-nitrotyrosine as an internal standard. The determinations of the methylated basic amino acids, the diastereoisomers of 5-hydroxylysine, and related compounds were carried out with concentrated 96-h hydrolysates (equivalent to 100-300 pg of protein/analysis) by the single-microcolumn 50 X 0.28 cm) system packed with Dionex DC-4A resin (Zarkadas et al., 1987b)so that peaks adequate for these components could be obtained. Determination of Protein Mass in Skeletal Muscle. Recoveries of these unique amino acids were calculated on the protein content of each 96-h hydrolysate determined by the procedure described by Horstmann (1979). According to this method, a mean residue weight (WE, pg/nmol) is calculated for the amino acids constituting the proteins in skeletal muscle samples as 20

WE = C ( a i b i ) i=l

(1)

where ai is the mole fraction of an amino acid i found in the analyzed aliquot and bi is the molecular weight of amino acid residue i as described by Horstmann (1979). A conversion factor F (pg/nmol) is used for determining the protein mass in each hydrolysate sample analyzed, which is the apparent mean residue molecular weight increased in proportion to the missing tryptophan (a%) and cys(e)ine (ac,) values from

F = W E / [ l - ( a n p+ ac,)l

(2)

The conversion factor F' (pg/mol) was also calculated according to eq 2, but for determining protein mass in the absence of tryptophan, cyst(e)ine, proline, and Pro(4-OH). These factors ( F and F? are constants characteristic for each protein or protein mixture. Thus, the amount of pg) in each hydrolysate can then be calculated protein (P, as 16

P = FCxj i=l

(3)

where x i are the nanomoles of each amino acid i found in the analyzed aliquot. Protein determination has also been carried out by the standard Kjeldahl procedure, and the results have been reported (Zarkadas et al., 1987a). (a)Determination of Connective Tissue Proteins. The amounts of the unusual protein-bound amino acids Lys(&OH) and Des, which occur exclusively in vertebrate connective tissue proteins, Le., collagens and elastin, have been used to determine the contents of these extracellular matrix proteins in three selected porcine skeletal muscles. In this chemical approach the distribution of elastic fibers

1134

Zarkadas et al.

J. Agric. Food Chem., Vol. 36,No. 6, 1988

Table I. Average Values for 5-Hydroxysine and 4-Hydroxyproline Computed from the Known Distribution of Types I, 111, and IV Collagens in Skeletal Muscles av no. res contrib by collagen av no. res/1000 total" collagen type distribn,b collagen type type standard chain assoc Pro(4-OH) Lys(5-OH) % total muscle collagen Pro(4-OH) Lys(5-OH) [CY1(I)lZ~2(I) 98 10 61.5 60.3 6.2 I I11 [a1(111)13 118 5 33.5 39.5 1.7 [a1(IV)1zCY2(IV) 120 42 5.0 6.0 2.1 IV LI3ni' 105.8 10.0 "Average values computed from the data of Miller and Cay (1982), Laurent et al. (1981), and Light (1985). (See also Table VIII.) bData taken from Light and Champion (1984) and Light et al. (1985).

(elastin) could be calculated from the amounts of Des found in the acid hydrolysates of these tissues and the total collagen and collagen-like proteins (Porter and Reid, 1978; Anglister et al., 1976) from the amounts of the aLys(5-OH) diastereoisomers present. Amorphous elastin (Foster, 1982),which was used as standard for comparison, contains 3 residues of Des/1000 total amino acids and has a mean residue weight (WE) of 85.06 (Table I). The anhydrous M,(i) of Des is 454.54. Light and Champion (1984) and Light et al. (1985) have shown that types I and I11 collagens accounted for 61 and 35% of the recovered collagen in the epimysium, perimysium, and endomysium of skeletal muscle, while type IV collagen accounted for the remaining 5%. Thus, a mean for the diastereoisomers of Lys(5-OH) content of n[ = 10.0 residues/1000 total amino acid residues in muscle collagen could be computed from the relative distribution of collagen types and their respective Lys(5-OH) contents presented in Table I. The average residue weight (WE) for collagen is 91.1, and each of the diastereomersof Lys(5-OH) has an anhydrous M, of 145.18. A method to calculate the amount of a specific protein in skeletal muscle is

Pj =

ci

[1OOO]

e

WE(Pj)

-.-

(4)

ni

This value is in close agreement with that reported by Etherington and Sims (1981). (b) Determination of the Amount of a Mixture of Two Proteins in Skeletal Muscle. The amount of two proteins (Pland Pz) containing the same or different amounts of the same unique amino acid (C, and C2) can also be determined from the total amount (C,) of that amino acid as described previously (Zarkadas et al., 1988): CT =

c1+ c2

(5)

By substituting C1 and C2 by the amounts of each protein from eq 4, the following relationship can be derived: 2

C(P1+ P2) =

j=l

PdP2 + 1 ni/WE(PJ + (P1/P2)n,'/WE(PJ

1OOOCT M,(d

(6)

The amount of the two proteins (P, and Pz) is dependent on the relative composition of the two proteins in the sample (Pl/P2), the number of residues of the unique amino acid (n;) per 1000 amino acid residues, and the total amount of the unique amino acid (CT). Equation 6 is analogous to eq 4 for one protein with a unique amino acid. As P1/P2 0, eq 6 will simplify to eq 4. (c) Determination of the Myofibrillar Proteins Myosin and Actin. Since sequence studies (Elzinga et al., 1973; Elzinga and Collins, 1977; Vandekerckhove and Weber, 1979; Tong and Elzinga, 1983; Maita et al., 1987) have shown that 1mol of actin (A) contains 1mol of His(r-Me) and that 1mol of myosin (M) contains 2 mol of His(r-Me), the quantitation of protein-bound His(.r-Me) can be used as an index for determining these two principal myofibrillar proteins in muscle tissues. Considerable care must be taken, however, to extract all soluble histidine dipeptides, especially balenine (Carnegie et al., 1982; Harris and Milne, 1987), prior to acid hydrolysis of muscle tissues. Thus, the total amount of His(r-Me) in the selected porcine tissues determined by the present method (Zarkadas et al., 198713) represents the sum of the distribution of His(r-Me) in the myosin and actin present in each of the porcine skeletal muscle tissue investigated and can be calculated according to +

where WE(P,) is the weight equivalent of a specific muscle protein j determined from eq 1as described by Horstmann (1959), n/ is the number of amino acid residues (i)/lOOO amino acid residues, and M,(i) is the anhydrous molecular weight of the unique amino acid i. The following analytical conventions, derived from eq 4, can therefore be used for calculating the amt of collagen (Pc)= amt of Lys(5-OH) X 63.3 (4a) and amt of elastin

(PE)

= amt of Des

X

62.4

(4b)

Thus, the total connective tissue content of porcine skeletal muscle (in grams per kilogram of total protein) could then be calculated from the sum of collagen (PG) and elastin (PE) found in these tissues. An alternative method for calculating the amount of collagen and elastin involves substitution into eq 6 below the total amount of Pro(4-OH) found in both proteins (Berg, 1982; Etherington and Sims, 1981; Rucker, 1982) from the known Pro(4-OH) contents of collagen (ni' = 105.8; see Table I) and elastin (ni = 22) as described previously (Zarkadas et al., 1988). The anhydrous molecular weight of Pro(4-OH) is 113.12. The relative distribution (Pc/PE)of collagen and elastin per unit of tissue can therefore be calculated either from the Lys(5-OH) and Des contents (eq 4a and 4b) or from the amounts of Pro(4-OH) found with eq 4c amt of connective tissue (Pc + P E ) = amt of Pro(4-OH) X 8.03 (4c)

CT = C A

+ CM

(5)

where CT is the total protein-bound His(r-Me) in myosin and actin (in grams per kilogram of the total protein), CA is the amount (9) of His(T-Me) in actin, and CM is the amount (g) of His(T-Me) in myosin in 1kg of total protein (Zarkadas et al., 1988). Since the relative amounts of actin (M,(A) 41 782; Elzinga et al., 1973) and myosin (M,(M) 521 000; Yates and Greaser, 1983) per unit of tissue (see eq 4), expressed as moles per kilogram of total protein, are related to the distribution of His(r-Me) in myosin and actin of skeletal

J. Agric. FoodChem., Vol. 36,No. 6, 1988

Quantitation of Contractile Protein in Porcine Muscle

1135

Table 11. Amino Acid (AA) Composition (Grams of Amino Acid per Kilogram of Protein) of Selected Porcine Skeletal Muscles Excised from Mature Yorkshire Sows mean f SEM" external amino acid sternomandibularis semitendinosus semimembranosus weighted CV" F 2.26 0.85" 89.48 f 0.47 90.88 f 0.48 90.47 91.06 f 1.43 aspartic acid 1.94 45.30 f 0.33'98 45.37 8.76* 43.93 f 0.328 46.87 f 0.77' threonine 41.12 1.65 14.01* 40.56 f 0.468 40.00 f 0.638 42.79 f 0.77' serine 2.46 1.26" 145.18 147.24 f 0.72 144.36 f 2.11 143.93 f 1.02 glutamic acid 4.78 7.85* 42.55 38.62 f 0.858 45.01 f 0.63' 44.03 f 1.12' proline 48.62 41.83 f 1.528 6.65 9.53* 51.40 f 0.67' 52.62 f 1.91' glycine 1.69 4.07' 58.44 56.88 f 0.42 59.28 f 0.35 59.17 f 0.58 alanine 2.94" 8.55 f 0.25 8.52 5.74 8.99 f 0.20 8.01 f 0.25 cysteine 1.87 7.54* 50.44 f 0.42' 49.66 48.11 f 0.558 50.44 f 0.68' valine 1.11" 40.59 42.35 f 1.42 14.01 42.77 f 0.85 36.64 f 5.84 methionine 10.26* 3.55 45.82 48.72 f 1.72' 43.03 f 0.678 45.72 f 0.64's isoleucine 1.74 78.72 79.45 f 0.61 1.34" 78.26 f 1.07 78.46 f 0.16 1eu cin e 2.5P 4.48 39.64 39.91 f 0.38 37.96 f 0.66 41.06 f 0.93 tyrosine 39.27 40.13 f 0.19 0.92" 6.82 40.07 f 0.07 phenylalanine 37.62 f 2.53 36.64 41.28 f 1.08' 15.54* 33.61 f 0.228 5.05 35.02 f 0.868 histidine 1.95" 1.52 87.99 87.06 f 0.20 88.56 f 0.49 lysine 88.34 f 0.94 68.88 8.98* 0.68 67.90 0.388 68.78 f 0.438 69.96 f 0.15' arginine 7.15 3.68" 8.69 7.43 f 0.32 7.64 f 0.17 tryptophan 6.38 f 0.10 11.46 11.36* 24.86 15.53 f 1.79' 5.12 f 1.168 13.73 f 0.72' 4-hydroxyproline 2.941 12.15 10.33* 3.28 f 0.150 2.015 f 0.028 5-hydroxylysine 3.53 i 0.41 0.221 31.48 1.64' 0.167 f 0.002 0.205 f 0.085 0.290 f 0.003 Ne-methyllysine , 0.227 25.22 6.82" 0.040 f 0.0 0.227 f 0.018 0.141 f 0.054 PJP-dimethyllysine 2.12" 0.971 54.12 1.35 f 0.16 0.589 f 0.05 NB,P,NB-trimethyllysine 15.42 2.408 f 0.43 6.48" 2.345 N"-methylhistidine 1.665 f 0.055 2.962 f 0 8.659 29.56 6.987 f 0.556 4.315 0.175 W-methylhistidine 14.674 f 0.50 8.83* 0.323 f 0 0.147 f 0.037 0.393 23.20 20.75* 0.715 f 0.105 unknown peak 17 14.19 f 1.30 5.82 2.24" ammonia 12.62 f 0.017 13.63 f 0 14.45 177.97 175.79 176.90 total AA Nb 677.15 f 15.31 613.85 f 24.94 total protein, g protein/ 513.58 f 9.54 kg dry w t EAA indexc 74.28 74.96 77.70 47.03 47.31 protein scorec 40.36 0.109411 0.109896 0.111470 WE? pglnmol 0.110767 0.11 1450 0.113028 F,d pg/nmol F': pg/nmol 0.118294 0.119442 0.118977 ~~~

~

*

*

" Mean values and standard error of measurements (SEM) for 3 replicates and 48 determinations. Significance: F, values from analysis of variance; **, P < 0.01; *, P < 0.05. Key: ns, not significant; CV, coefficient of variation. (f, g) Means along a horizontal column with different superscripts are significantly different. *Calculated according to Heidelbaugh et al. (1975). From Oser (1951) and Block and Mitchell (1946). dThe WE and F constants calculated according to Horstmann (1979). eThe conversion factor F'calculated according to eq 2 but for determining protein mass in the absence of tryptophan, cyst(e)ine, proline, and 4-hydroxyproline. muscle, the amounts of myosin (PM) and actin (PA) in porcine muscles (grams per kilogram of total protein) can thus be calculated from the total amount (CT) of His(7-Me) from eq 5 and the molar ratio of actin and myosin by the following equation: 2

A

[

41782

M

151.2+--A + 2M

521000 ]CT (6) 151.2

Substituting the molar ratio of actin to myosin reported by Murakami and Uchida (1985) ( A / B = 6) in eq 6, the sum of actin and myosin in porcine skeletal muscles can be made as 2

(PA j-1

+ P M ) = [207 + 431]cT

(64

or 2

C ( P A+ P M )= 638CT

j=l

(6b)

Yates and Greaser (1983) have shown that the sum of actin and myosin in the myofibril accounts for 65% of the total myofibrillar protein by weight. Thus, the total

myofibrillar protein in grams per kilogram of total protein in porcine skeletal muscle can also be calculated as 2

amt of myofibrillar protein =

j=1

( P A + PM) 0.65

CT

= 981CT

(7) Statistical Analysis. Data processing and linear regression analysis of the results were carried out by a Fortran computer program developed for this purpose. Analysis of variance conducted on the amino acid data for a completely randomized block design (factorial) was carried out by the Statistical Analysis System (SAS, 1982) general linear model procedure. RESULTS AND DISCUSSION

Amino Acid Composition of Selected Porcine Muscles. The overall amino acid compositions of the selected external sternomandibularis, semitendinosus, and semimembranosus porcine skeletal muscles excised from mature Yorkshire sows, usually being marketed for meat processing, and levels of statistical significance obtained from analysis of variance are summarized in Table 11. The results expressed as grams of amino acid residues per kilogram of total protein show deviations of less than 100 f 3.0% from the weighted mean between three animals within the same treatment. The main advantage of this

1136

J. Agric. Food Chem., Vol. 36,No. 6, 1988

Zarkadas et al.

Table 111. Recoveries of the Intracellular and Extracellular Matrix Protein Fractions (F1 and F2) Isolated from Three Selected Porcine Skeletal Muscles by the Combined Methods of McCollester (1962) and Laurent et al. (1981) mean f SEM" F1 intracellular SDS-sol muscle F2 extracellular SDS-insol protein protein fraction fraction g dry matter/ g protein/ g dry matter/ g protein/ porcine skeletal muscle 10.0 g wet tissue 100 g proteinb 10.0 g wet tissue 100 g proteinb 89.78 f 0.80 0.1689 f 0.02 10.22 f 1.21 external sternomandibularis 2.3517 f 0.021 semitendinosus 2.2806 f 0.009 94.55 f 0.373 0.1573 f 0.021 5.45 f 0.073 94.80 f 0.007 0.1729 f 0.007 5.20 f 0.211 semimembranosus 2.2632 f 0.17

" Mean f standard error of measurement (SEM) for three replicates. bProtein content was determined by amino acid analysis according to Horstmann (1979). unit of expressing the composition of a protein mixture is that it reflects the relative amounts of the amino acids present (Tristram and Smith, 1963; Eastoe, 1967) since the influence of fat, ash, and moisture is eliminated. Protein determinations were carried out in each acid hydrolysate as described previously (Horstmann, 1979; Nguyen et al., 1986). This method is based upon knowledge of the amino acid composition of the protein or protein mixture and yields accurate estimates of the amount of protein present calculated by eq 1-3. The constants, weight equivalent (WE, pg/nmol) and conversion factors (Fand F', pg), for each of the porcine muscles (Table 11) or muscle protein fractions (Tables IV and V) investigated have been determined and can be used in all subsequent quantitations of these muscle tissues following standard procedures as described by Horstmann (1979), Peterson (1983), and Nguyen et al. (1986). A comparison of the amino acid profiles of the porcine external sternomandibularis, semitendinosus, and semimembranosus muscles investigated shows that although many of the individual amino acid values appeared to be very similar; muscle to muscle variations in amino acid content from the weighed mean were significant with respect to 12 amino acids among these muscles. The variation noted for threonine, serine, proline, glycine, and valine among the three muscles evaluated was statistically significant (P < 0.05), with the external sternomandibularis muscle being consistently higher in these amino acids than either of the other two muscles. Mean proline values ranged from 48.13 g/kg of protein in sow semitendinosus to 41.6 g/kg of protein in semimembranosus for sows. The mean value obtained for arginine in the semitendinosus muscle was significantly higher (P < 0.05) compared to that found in the other two muscles. The weighted mean values obtained for the amino acid composition of the three muscles investigated are consistently higher compared to those obtained for the semimembranosus muscle. In all the tissues studied, Pro(4-OH) and Lys(5-OH) varied significantly between porcine muscles, with the semimembranosus muscle containing the lowest mean Pro(4OH) and Lys(5-OH) values. The variation noted in the His(T-Me)content among the three untreated porcine muscles evaluated (Table 11) was significant (P < 0.05), with the semimembranosus muscle being much higher in total His(T-Me), 14.67 g/kg of protein, compared to that found in the semitendinosus (6.99 g/kg of protein) and external sternomandibularis (4.3 g/kg of protein) muscles. Skeletal muscles are known to contain variable amounts of soluble histidine dipeptides including balenine (Carnegie et al., 1982; Harris and Milne, 1987), /3-alanyl-L-W-methylhistidine. These dipeptides must be extracted from muscle tissues prior to acid hydrolysis. One major limitation of the muscle tissue extraction procedures (Rangeley and Lawrie, 1976; Bligh and Dyer, 1959; Happich et al., 1984; Carnegie et al., 1984), however, is that

often high incidences of unemptied muscle cell segments are encountered. This emptying phenomenon, although not fully understood (McCollester, 19621, has been related to irreversible contraction of muscle cells during homogenization (McCollester and Semente, 1964). Thus, conventional methods of cellular disruption are unable to break down contracted muscle cell cytoskeletal membranes, i.e., endoplasmic reticulum, Z-bands, etc., to produce a satisfactory dissolution and separation of the intracellular soluble structures from the extracellular matrix insoluble proteins. Unique Basic Amino Acid Content of Porcine Intracellular and Extracellular Protein Fractions (F1 and F2). To quantitatively establish the occurrence and variation of protein-bound His(T-Me) in porcine skeletal muscles, the combined procedures of McCollester (1962) and Laurent et al. (1981) have been used to effectively separate the intracellular SDS-soluble muscle protein fraction (Fl) from the extracellular SDS-insoluble protein fraction (F2). Each of these fractions (F1 and F2) was subjected to detailed amino acid analysis and protein determination by the use of the chromatographic methods (Zarkadas et al., 1986,198713) developed to quantitate the unique amino acids that occur in those proteins. The yields obtained for F1 and F2 are summarized in Table 111. The total yield obtained for the intracellular muscle protein fraction (Fl) in the porcine external sternomandibularis muscle averaged 89.8% compared to 96.6% and 96.8% found in the semitendinosus and semimembranosus muscles, respectively. The external sternomandibularis muscle contained a higher concentration of extracellular matrix proteins (10.2%) compared to that found in the semitendinosus (5.45%) and semimembranosus (5.2%). Since recoveries for each of the fractions have been calculated on total protein determined from their resepective amino acid composition, as presented in Tables IV and V, the average protein yields reported in Table I11 represent accurate estimates of the amount of protein present. These results are comparable to those reported by Bendall (1967) and Dransfield (1977) for bovine muscles. Typical chromatographic separations obtained, when samples of the intracellular and extracellular porcine muscle protein fractions (F1 and F2) were analyzed on a 50 x 0.28 cm column of Dionex DC-4A resin with two 0.35 M sodium citrate buffers, pH 5.701 and 4.501 (buffer changed at 215.7 min), at two column temperature (28 "C changed to 73 "C a t 294.7 min), are illustrated in Figure 1. As may be seen in Figure l A , the analysis of an intracellular muscle (external sternomandibularis) protein fraction (Fl) 96-h hydrolysate (100-pL equivalent to 200 pg of protein/analysis) by the present method (Zarkadas et a]., 1987b) revealed the complete separation, at pH 5.700, of both methylated lysines along with four as yet unidentified ninhydrin-positive peaks, designated 2, 7, 8, and 10. The Arabic number assigned to each unknown

Quantitation of Contractile Protein in Porcine Muscle

J. Agric. FoodChem., Vol. 36, No. 6, 1988

1137

Table IV. Comparison of the Amino Acid Composition (Grams of Amino Acid per Kilogram of Protein) of SDS-Solubilized Intracellular Muscle Protein Fractions (Fl) Isolated from Three Typical Porcine Skeletal Muscles of Mature Yorkshire sows porcine skeletal msucle SDS-sol muscle protein fraction (Fl) external signif level sternomandibularis semitendinosus semimembranosus (between ( N = 4) ( N = 4) (N = 4) wt mean muscles)b meanfSEMa CV meanfSEMa CV meanf SEM" CV ( N = 12) CV F 0.63 98.33 f 0.22' aspartic acid 96.16 f 0.31' 0.44 96.63 f 0.608 1.22 97.71 0.83 5.47* threonine 48.48 f 0.38' 1.58 47.04 f 0.41' 1.74 48.59 f 0.369 1.50 48.04 1.67 6.07* 3.58 36.04 f 0.27' 42.75 f 0.76' 1.47 41.82 f 0.708 serine 3.37 40.20 3.08 35.92** glutamic acid 163.38 f 0.49 0.60 162.82 f 0.40 0.50 164.07 f 0.64 0.79 163.42 0.64 2.63" 37.95 f 1.19 6.27 36.94 f 2.90 2.95 37.13 f 0.44 2.33 proline 37.34 4.27 0.60M glycine 30.18 f 0.119 0.69 31.04 f 0.08' 0.53 29.54 f 0.11' 0.72 30.25 0.65 52.66** alanine 50.71 f 0.15' 0.60 51.14 f 0.20' 0.77 49.99 0.289 1.12 50.61 0.85 6.48** cysteine 8.33 f 0.078 1.55 10.15 f 0.14' 2.75 8.14 f 0.018 0.22 8.87 2.00 150.88** 55.44 f 0.11 0.39 56.07 f 0.33 1.19 56.66 f 0.47 56.06 1.21 2.6gM valine 1.68 methionine 28.83 f 0.23 1.54 28.83 f 0.09 0.59 28.64 f 0.23 28.80 1.57 1.32 O.5gM isoleucine 53.35 f 0.23 0.86 54.47 f 0.32 1.17 54.16 f 0.36 0.37 53.99 1.15 2.20M 88.11 f 0.18 0.41 87.99 f 0.36 leucine 0.83 88.01 f 0.27 0.60 88.04 0.64 0.88M tyrosine 40.68 f 0.398 1.94 44.13 f 1.51' 6.82 40.22 f 0.658 41.68 4.66 4.52* 3.22 phenylalanine 44.00 f 0.40 1.81 44.61 f 0.60 2.67 44.62 f 0.84 3.80 44.41 2.88 0.18M 27.66 f 0.208 1.43 29.08 f 0.36' histidine 2.53 27.88 f 0.368 2.62 28.21 2.27 4.85* lysine 95.00 f 1.67' 1.39 91.85 f 0.679 1.46 96.46 f 0.41' 0.86 1.26 17.74** 94.44 68.94 f 0.338 0.97 68.77 f 0.548 arginine 1.58 70.59 f 0.26' 0.72 69.43 1.14 6.63* tryptophan 16.98 f 0.038 0.29 19.38 f 0.09' 0.91 15.52 f 0.03' 0.32 17.29 0.63 232.28** ornithine 0.31 f 0.003' 2.09 0.264 f 0.0058 3.51 0.266 f 0.018 9.41 0.284 5.70 8.68** P-methyllysine 0.10 f 0.0018 3.95 0.095 f 0.0018 2.53 3.73 0.101 0.109 f 0.002' 3.51 15.36** P,Pfl-trimethyllysine 0.41 f 0.01' 5.26 0.350 f 0.007' 3.97 0.453 f 0.01' 5.41 0.405 5.07 25.63** Nl-methylhistidine 0.592 f 0.009' 3.16 0.589 f 0.01' 3.47 0.521 f 0.0059 2.07 0.568 3.02 22.16** ammonia 15.30 f 0.948 12.26 10.58 f 1.03' 19.50 25.01 f 1.78' 14.24 16.96 15.39 31.77** total AA NC 172.53 168.63 180.82 WE? pg/nmol 0.11262 0.11295 0.11278 0.120256 0.120944 0.120170 F,d pg/nmol 0.121176 0.120428 F': ,ug/nmol 0.120508

*

Mean values and standard error of measurements (SEM) for four replicates. *Significance (N = 12): **, P C 0.01; *, P C 0.05. Key: ns, not significant; CV, coefficient of variation. (f, g, i) Means along a horizontal column with different superscripts are significantly different. 'Calculated according to Heidelbaugh et al. (1975). "Calculated according to Horstmann (1979) and Nguyen et al. (1986), with eq 1 and 2. 'The conversion factor F'was calculated according to eq 2 but for determining protein mass in the absence of tryptophan, cyst(e)ine, proline, and 4-hydroxyproline. PORCINE EXTERNAL STERNOMANDIBULARIS A. MUSCLE PROTEIN FRACTION lF1I

ACIDIC AND

0.25

-

Unknown

w

0

z

g2

I

I

I

I

1

I

0.50-

a

6. CONNECTIVE

0.25

Unknowns

-

FRACTION (F2)

d0

lb

1 Bo

2& TIME (min)

280

3bo

380

F i g u r e 1. Typical elution patterns of the methylated basic amino acids and related compounds in the intracellular and extracellular protein fractions from the external sternomandibularis porcine skeletal muscle on an 0.28 X 50 cm analytical microcolumn of Dionex DC-4A resin. The curves show absorbance at 570 nm. (A) Separations of all methylated lysines and histidines in the intracellular muscle protein fraction (Fl). (B)Analysis of a 96-h hydrolysate of the extracellular matrix protein fraction (F2). Key: Tyr(N02), 3-nitrotyrosine; Lys(5-OH), 5-hydroxylysine; aLys(bOH), allo-5-hydroxylysine; Lys(Me), I@-methyllysine; Lys(Me,+), NB,@,IP-trimethyllysine; His(r-Me), N'-methylhistidine.

peak in consecutive order indicates its relative elution time from the microcolumn, and the results calculated relative to the internal standard [Tyr(N02)]are presented in Table VI. Since the positions of the methylated histidines in the chromatogram are very sensitive to pH, it was found that, at pH 4.501 and 71 "C,it has been possible to accomplish the complete separation of His(7-Me) from an unknown ninhydrin-positive peak, designated 17 (Table

VI). This unknown compound (17) has also been found in variable amounts in other proteins and tissues. Analysis of an acid hydrolysate (96 h) of the extracellular matrix protein fraction (F2) from the external sternomandibularis porcine muscle (Figure 1B) revealed the complete separation of the diastereoisomersof Lys(&OH) along with five as yet unidentified stable components, designated 2, 7, 8, 10, and 11 (Table VI). Although pre-

J. Agric. Food Chem., Vol. 36,No. 6, 1988

1138

Zarkadas et al.

Table V. Amino Acid Composition (Grams of Amino Acid per Kilogram of Protein) of SDS-Insoluble Muscle Connective Tissue Fractions (F2) Isolated from Porcine Skeletal Muscles of Mature Yorkshire Sows porcine skeletal muscle SDS-insol connective tissue fractions (F2) signif levels external (between wt mean muscles)b sternomandibularis semitendinosus semimembranosus ( N = 12) CV F meanf SEM" CV meanf SEM" CV meanf SEM" CV asDartic acid 63.45 f 0.528 1.65 65.44 f 1.04' 3.19 67.43 f 0.49' 1.42 65.44 2.27 17.83** 20.24 f 0.35' 3.44 thieonine 21.34 f 0.268 2.48 25.97 f 0.94' 7.27 22.52 5.32 27.60** 39.18 f 0.89g 4.53 serine 40.68 f 0.33 46.36 f 1.85' 1.65 7.99 42.07 5.68 11.59** 1.05 111.37 f 2.46' glutamic acid 109.23 f 0.578 4.25 115.53 f 0.96' 1.67 112.04 2.75 10.04** 119.59 f 1.18' 1.97 107.87 f 2.288 proline 4.25 99.23 f 0.93' 1.88 2.94 108.90 30.69** glycine 0.67 183.95 f 4.14' 189.27 f 0.63' 4.51 167.17 f 0.718 0.85 180.13 2.79 21.03** alanine 79.87 f 1.38'~s 3.46 77.66 f 1.47' 74.24 f 0.509 3.79 1.36 77.26 3.12 5.48** 3.30 f 0.12' 7.56 cysteine 4.43 f 0.11' 5.13 2.14 3.87 f 0.048 3.87 44.11** 5.16 29.16 f 0.548 valine 29.34 f 0.808 3.75 5.49 32.67 f 0.36' 2.25 30.39 3.99 12.30** 9.99 methionine 9.16 f 0.458 10.74 f 0.47' 8.90 10.17 f 0.73'98 15.01 9.97 11.58 3.29* isoleucine 16.60 f 0.27' 3.28 18.44 f 0.428 4.51 21.99 f 0.14' 1.31 19.01 3.18 91.73** 2.48 leucine 34.37 f 0.42' 36.23 f 0.718 3.92 40.94 f 0.31' 1.58 37.18 2.79 50.82** tyrosine 12.79 f 0.958 12.45 f 0.818 14.83 13.06 16.55 f 1.00' 12.14 13.93 5.97* 13.25 phenylalanine 23.65 f 0.558 4.69 23.71 f 0.388 3.23 26.05 f 0.37' 2.85 24.47 11.44** 3.61 histidine 7.73 f 0.14' 3.71 8.72 f 0.318 7.08 10.83 f 0.34' 6.32 9.09 34.37** 6.17 lysine 37.94 f 0.62' 3.29 4.18 39.43 f 0.868 43.88 f 0.60' 2.73 40.42 24.42** 3.43 arginine 82.98 f 1.20 2.89 3.67 79.69 f 1.46 80.74 f 0.61 1.51 81.14 0.28"# 2.84 4-hydroxyproline 111.50 f 1.878 3.37 118.12 f 2.19' 3.70 107.10 f 1.688 3.14 112.24 16.23** 3.44 5-hydroxylysine 9.11 f 0.058 9.81 f 0.06' 1.04 1.16 8.64 f 0.07' 1.60 9.19 1.27 101.01** 0.224 f 0.0038 isodesmosine 0.291 f 0.00' 2.75 0.58 0.185 f 0.001' 2.04 0.234 1.84 620.28** 0.225 f 0.0028 desmosine 0.291 f 0.002' 2.00 0.194 f 0.001' 1.83 1.80 0.237 1.90 486.86** ornithine 0.362 f 0.0048 0.458 f 0.004f*8 1.91 0.588 f 0.098' 32.91 2.25 0.472 24.14 4.45* ammonia 24.61 f 7.53 61.65 32.86 f 6.77 41.25 70.21 1.29"# 53.00 f 19.93 36.76 75.19 199.28 total AA N' 203.55 219.07 WE,d rg/nmol 0.093110 0.93726 0.095323 0.093389 0.094105 F,d ,ug/nmoI 0.095665 F',e pg/nmol 0.117781 0.118034 0.117862 "Mean values and standard error of measurements (SEM) for four replicates. "Significance: **, P < 0.01; *, P < 0.05. Key: ns, not significant; CV, coefficient of variation. (f, g, i) Means along a horizontal column with different superscripts are significantly different. eCalculated according to Heidelbaugh et al. (1975). dCalculated according to Horstmann (1979) and Nguyen et al. (1986), by eq 1 and 2. 'The conversion factor F' calculated according to eq 2 but for determining protein mass in the absence of tryptophan, cyst(e)ine, proline, and 4-hydroxyproline. Table VI. Elution Times and Contents (Nanomoles per Milligram of Protein) of Unknown Ninhydrin-Positive Compounds Separated from Selected Porcine Skeletal Muscle Protein and Connective Tissue Fractions porcine skeletal muscles signif levels external (between unknown elution sternomandibularis semitendinosus semimembranosus muscles)b F comDd time,min meanf SEM" CV meanf SEM" CV meanf SEM" CV wtmean CV SDS-Solubilized Muscle Protein Fraction F1 0.838 f 0.126d 29.98 0.983 f 0.33d 6.82 1.243 38.18 1.907 f 0.39' 40.87 5.99* 2 (77) 7 (137) 0.783 0.21' 53.14 0.139 f 0.007d 9.61 0.367 f 0.025d 13.58 0.430 56.26 7.26* 1.968 f 0.075d 7.67 2.958 f 0.255d 17.24 3.488 19.91 28.19** 5.538 f 0.54' 19.48 8 (146) 4.40 22.72 2.93 26.310 f 0.578' 4.79 63.34** 10 (211) 17.900 f 0.66e 7.36 23.953 f 0.35d 4.22 0.914 f 0.08d 18.75 1.283 f 0.125' 19.52 0.98 17.95 9.69** 17 (350) 0.747 f 0.016d

*

2 7

(77) (137)

10

(211)

12

(267)

SDS-Insoluble Muscle Connective Tissue Fraction F2 0.640 f 0.031d 2.31 1.143 f 0.058c 10.19 0.536 f 0.053d 19.92 2.684 f 0.032' 2.31 0.821 f 0.244d 59.31 8.72 2.239 f 0.098c 2.995 f 0.038d 2.19 4.810 f 0.814' 33.84 3.802 f 0.326c*d 17.16 5.071 f 0.577C3d 22.75 8.35 3.824 f 0.246d 12.85 6.199 f 0.258'

0.77 1.92 3.87 5.02

12.67 15.93 26.18 15.58

43.88** 40.70** 3.22"s 9.06**

"Mean values and standard error of measurements (SEM) for 3 replicates and 48 determinations. bSignificance( N = 12): **, P < 0.01; *, P < 0.05. Key: ns, not significant; CV, coefficient of variation. (c-e) Means along a horizontal column with different superscripts are significantly different.

vious work from this laboratory showed that the determination of Lys(5-OH) can be made from the sum of the values obtained from its diastereoisomers after epimerization in 6 M HCl at 110 "C for 96 h (Zarkadas, 1975), as shown in Figure 1B unknown peak 8 partially coelutes with the Lys(5-OH) diastereoisomer, even after epimerization, thus interfering with its quantitation. For this reason, only the aLys(5-OH) diastereoisomer after epimerization (96 h) is now being routinely used in the quantitation of this important amino acid found in collagen and collagen-like proteins (Reid, 1982; Anglister et al., 1976), as well as in the extracellular matrix and muscle tissues.

A summary of the amino acid composition of the SDSsolubulized intracellular muscle protein fractions (Fl) isolated from three typical porcine skeletal muscles of mature Yorkshire sows, and levels of statistical significance obtained from analysis of variance, is presented in Table IV. Although the amino acid profiles (grams per kilogram of protein) of the intracellular protein fractions from external sternomandibularis, semitendinosus, and semimembranosus porcine muscles are very similar, characteristic differences were noted in their contents of aspartic acid, threonine, serine, glycine, alanine, cyst(e)ine,tyrosine, histidine, lysine, arginine, and tryptophan. The variation

(kantitaUm of cOmn*le Protein in Pordno Muscle

J. A@.

3

4

5

6

FwdCnem.. Vd. 36,No. 6, 1988

7

llS@

8

pislve 2. comperimn of onedimensional SDBPAGEon 10% polyaaylamide sleh gel eleetrophoretcseparations of Gactin monomer8 purified from s e l d muscle tissues LU) outlined in the puriiication scheme in Materials and Methods. The G-actin monomers were reduced with dithiothreitol, denatured with SDS eample bufferat 100 O C for 5 min, and subjected to eleetrophorssis(Weber and Oaborn, 1975). The gels were then stained for protein with Coomassie Brilliant Blue C-250 and destained (Weber and Oabom,1975). Lanes: 1,standards [A. phosphorylase (92500 Da); B, bovine serum albumin (67000); C, ovalbumin (40000);D, carbonic anhydrase (30000)1; 2. chicken breast actin; 3, chicken leg muscle actin; 4, porcine semimembranosus actin; 5, porcine cardiac muscle actin; 6, bovine semimembranosus actin; 7, bovine cardiac muscle actin; 8, standards.

noted for serine, glycine, alanine, eyst(e)ine, lysine, and tryptophan among the three musclea evaluated wan highly significant (P< 0.00, with the semitendinosus muscle being consistently higher in total glycine, alanine, c y s t (e)ine, and tryptophan than either of the other two muscles. The external sternomandibularis and semimembranosus muscles contained the highest levels of lysine but had the lowest mean values of histidine. Variations. however, due to 36 determinationsbetween three replica& within the same muscle were small, indicating that analytical errors were also small relative to biological variability observed between muscles. Thus, the data presented in Table IV defined variability in amino acid content of the intracellular muscle protein fractions (Fl) among selected muscles. A comparison of the amino acid data presented in Tahlea I1 and IV indicate that practically all of the His(*-Me), 17.8-36.1% of total histidine, and about 93% of the total His(r-Me) found in porcine skeletal muscles were present as free or as soluble histidmebound dipeptides, which on acid hydrolysis yield &alanine, histidine, and His(*-Me). These results are in accord with those reported by Harris and Milne (1981, 1987) and Carnegie et al. (1982) for porcine muscles. These authors have shown that in young animals balenine constitutes more than 90% of the total nonprotein-bound His(r-Me), reaching 99.8% of the total His(r-Me) in older animals. Although the physiological function of anserine and balenine in skeletal muscle tissues has not been established, both p-alanine and eamosine are now beiig considered as neurotransmitters (Griffth, 1986). The values obtained for protein-bound His(r-Me) in the intracellular muscle protein fractions (Fl) of the selected porcine muscles (Table IV) show high reproducibility and low coefficients of variation, and within the precision of the chromatographic method (100i 2.5%), recoveries were found to be quantitative. These results show that while the protein-bound His(r-Me) contents of the external sternomandibularis (0.592 g/kg of protein) and semiten-

dinosus (0.589 g/kg of protein) " l e a appeared to be very similar, there was a small but statistically significant variation (P< 0.01) in the protein-bound His(r-Me) content (0.521 g/kg of protein) of the semimembranosus muscle. While most of this variation may be attributed to the presence of variable amounts of fast (type IIA), slow (type I), or mixed fast/slow (types IIA and IIB) classes of myosin heaw chains. the factors reswnsible for the small differences ih His(r-Me) contents among porcine skeletal muscles are less clear. The distribution of heavy myosin chain isoforms among different muscle fiber types with fast, mixed fast/slow, and slow properties has been shown to vary in a musclespecific manner (Gauthier and Lowey, 1979; Beecher et al., 1968; Beerman et d,1978; Suzuki and Camens, 1980. Kiessling et aL, 1982; Kieasling and Hanson, 1983; Lowey, 1986a,b; Mahdavi et al., 1986; Staron and Pette, 1987a,h). "-Methylhistidine Content of Porcine Semimembranosus Actin. To establish whether this small variation (11.9%) in protein-bound His(r-Me) among porcine muscles could be attributed to myosin heavy chain or actin isoforms, the His(r-Me) content of purified G-actin from sow's semimembranosus muscles was analyzed by the present method (Zarkadas et al., 1987b). The purification of actin was carried out by the procedure of Spudich and Watt (1971), followed by the polymerization and depolymerization steps recommended by Pardee and Spudich (1982). as described in Materials and Methods. The purified Gactin eluted from the Sephacryl G200 column (2.6 X 95 cm) with the depolymerization buffer of Pardee and Spudich (1982) was subjected to o n e and t w d i e n s i o n a l SDS-polyacrylamide gel electrophoresis (Weber and Osborn, 1975; Laemmli, 1970; OFarrell, 1975), and a single hand was observed after staining the gels for protein (Figure 2)8, demonstrating that the G-actin obtained by this procedure is essentially homogeneous (Pollard and Cooper,1986). As may be seen in Figure 2, similar single bands were observed with actins purified from bovine

1140

J. Agric. Food Chem., Vol. 36, No. 6, 1988

Zarkadas et al.

Table VII. Amino Acid Composition of Porcine G-Actin from the Semimembranosus Muscle mean, nmo1/1000 mole pg protein fraction" res noeb amino acid (c; + SEM)" (ai) (nif SE) kbedtnib aspartic acid 899.36 f 9.7 0.0944 35.43 f 0.38 35.20 threonine 699.33 f 14.1 0.0702 26.37 f 0.55 26.20 541.13 f 15.7 0.0568 serine 21.32 f 0.62 21.18 glutamic acid 1062.09 f 11.3 0.1114 41.84 f 0.44 41.57 proline 484.82 f 11.1 0.0509 19.10 f 0.44 18.97 750.77 f 9.8 0.0788 glycine 29.58 0.38 29.38 alanine 745.51 f 8.4 0.0782 29.37 f 0.33 29.18 119.41 f 1.0 0.0125 cysteine' 4.70 f 0.39 4.67 valine 563.27 f 12.5 0.0561 22.19 f 0.49 22.04 methionine 382.12 f 1.0 0.0401 15.05 f 0.04 14.95 isoleucine 734.89 f 11.8 0.0771 28.95 f 0.46 28.76 leucine 639.60 i 11.2 0.0671 25.20 f 0.44 25.03 tyrosine 409.07 f 6.0 0.0429 16.11 f 0.24 16.01 phenylalanine 267.68 f 6.3 0.0281 10.55 f 0.25 10.47 lysine 457.24 f 23.7 0.0480 18.01 f 0.93 17.89 histidine 198.23 f 10.9 0.0208 7.81 f 0.43 7.76 arginine 462.31 f 9.9 0.0464 18.21 f 0.39 18.09 tryptophanc 119.41 f 1.0 0.0125 4.70 f 0.04 4.67 W-methylhistidine 24.52 f 0.6 0.0026 0.97 f 0.02 0.96 total 9560.76 1.oooo 375.46 WE," pg/nmol 0.11125 F,pg/nmol 0.11441 MW

*

nearest integer ratiob (Ii) 35 26 21 42 19 29 29 5 22 15 29 25 16 10 18 8 18 5

I

340 27 22 39 19 28 29 5 21 16 30 26 16 12 19 8 18 5 1

373

375c

41 648

41 872c

The protein concentration, mole fraction (ai), weight equivalent (WE), and conversion factor (F)for G-actin from porcine semimembranosus skeletal muscle were computed according to Horstmann (1979). bDetermination of the residue number (ni SE), the goodness of fit (f) to integral amino acid residue numbers (Zi) as a function of the scaling factor ( k k t ) , equivalent to a molecular weight of 41 648 for G actin, was carried out by the computer-assisted procedures of Black and Hogness (1969) and Hoy et al. (1974). 'Data taken from Vandekerckhova and Weber (1978). Table VIII. Comparison of the Amino Acid (AA) Composition (Number of Amino Acid Rssidues/1000 Total Amino Acid Residues) of Extracellular Matrix Protein Fractions (F2)from Selected Porcine Skeletal Muscles with Mammalian Collagens and Avian Elastin porcine musclea intramuscular connective tissue fraction avian bovine external semisemicollagen stdbve elastin semitendinosus elastind amino acid sternomandibularis tendinosus membranosus t m e I t m e I11 t m e IV std' (insol) 51.33 53.29 55.84 43 42 46 2 7.4 aspartic acid 8.2 18.64 19.70 24.49 18 13 29 3 threonine 50.74 8.3 39.18 43.77 33 39 35 5 serine 17.1 78.78 80.85 85.30 71 71 74 12 glutamic acid 110.9 97.41 120 107 114.68 104.12 81 128 proline 279.07 335 350 331 301.94 352 325.9 g1y cine 308.63 111 96 36 99.53 102.37 176 223.1 104.59 alanine 3.58 1 1 4.03 2.98 cysteine 31.42 26 14 31 175 145.8 27.75 27.40 valine 7.27 6 8 15 7.67 6.50 methionine 18.52 9 13 34 19 26.5 15.27 isoleucine 13.65 47 61.4 23 22 53 34.47 30.0 28.27 leucine 9.67 12 8.4 2 3 6 7.15 7.30 tyrosine 16.87 29.8 23 12 8 30 15.10 14.96 phenylalanine 1 7.52 0.6 6 6 6 5.96 5.25 histidine 5.3 4 23 30 6 32.63 28.83 27.56 lysine 49.27 5.7 5 50 46 29 47.82 49.46 arginine nd 1 1 nd nd 3-hydroxyproline 22 12.4 97.89 90.27 98 118 120 91.79 4-hydroxyproline 5.71 10 5 42 6.40 5.88 5-hydroxylysine 1.2 0.0600 0.0388 3 0.0459 desmosine 2.1 0.0407 3 0.0600 0.0461 isodesmodine 0.15 0.48 0.52 0.48 0.45 0.48 0.44 Pro(4-OH)/(Pro + Pro(4-OH)) 0.27 0.17 0.15 0.17 0.15 0.18 Lys(5-OH)/(Lys + Lys(5-OH)) OMeans of 4 replicates and 64 determinations, present study. *Average values recalculated from the data of Miller and Gay (1982), Light and Champion (19841, and Light (1985). CDatafor amorphous elastin taken from Foster (1982). dData for bovine semitendinosus muscle elastin taken from Bendall (1967). OThe average WE values for muscle collagen and connective tissue calculated according to Horstmann (1979): Ei.ls(WE muscle collagen) = [(0.615 X 91.14) + (0.335 X 90.90) + (0.05 X 95.94)] = 91.10 g/mol (see also Table I); Cilt(WE muscle connective tissue) = [(0.0473 X 85.06) + (0.9527 X 91.10)] = 90.014 g/mol (see also Table I).

semimembranosus muscle and bovine and porcine cardiac muscles as well as from chicken breasts or leg muscles. The molecular weights of G-actin purified from porcine and bovine semimembranosus skeletal muscles were deter-

mined as 41 648 and 41 279, respectively. These results are in accord with those reported by Vandekerckhove and Weber (1978; 1979). The mean values and standard errors of the amino acid

J. Agric. Food Chem., Vol. 36,No. 6 , 1988

Quantitation of Contractile Protein in Porcine Muscle

1141

A. Standard Calibration A. A. Mixture

0.500

0.250

w

0 2

I

I

I

I

t

I

--

a

H i s ( Me) ~

unknown 3

0.250

-

0

b

Orn

m

I

I

I

I

I

1

60

120

180

240

300

360

TIME (min)

Figure 3. Chromatographic separation of all methylated basic amino acids and related compounds of actin from the porcine semimembranosus skeletal muscle. (A) Separation of a synthetic amino acid calibration mixture. (B) Typical separation of a 96-h hydrolysate of porcine semimembranosus actin. The curves show absorbance at 570 nm. Key: Tyr(NOz),3-nitrotyrosine; GlcN, glucosamine; Lys(BOH),5-hydroxylysine;aLys(5-OH),ah-5-hydroxylysine;Lys(Me),M-methyllysine;Lys(Mez),i@,I@-dimethyllysine;Lys(Me,+),

M,NB,M-trimethyllysine;His(r-Me),W-methylhistidine. composition of G-actin purified from porcine semimembranosus are presented in Table VII. The data are expressed as nanomoles/1000pg of protein and represent the average values of triplicate determinations obtained from duplicate 24-, 48-, 72- and 96-h hydrolysates. The values for only two amino acids, cysteine and tryptophan, are taken from Vandekerckhove and Weber (1978). The protein concentration of individual hydrolysate samples was determined by the procedure described by Horstmann (1979). The average weight equivalent (WE = 0,11125 pg/nmol) and conversion factor ( F = 0.114 41 pg/nmol) obtained are also listed in Table VII. The large number of determinations used to obtain the data listed in Table VI1 were necessary to ensure low coefficients of variation (2%) for any given amino acid residue so that the computer-assisted method of Hoy et al. (1974) could be applied for determining both the nearest integer ratios (i.e., frequencies) of amino acid residues and a minimum molecular weight for this protein. The results obtained (ni SE) by this method are given in Table VII. Since the values of niare seldom integral numbers, the ni values obtained for G-actin from the porcine semimembranosus muscle were multiplied by a series of values for the k scaling factor ranging from 0.91 to 1.1 in steps of 0.0001 with 20000 points available for plotting in the search for kbest = 0.9934. A good fit was indicated by a value for f = 0.4905. The nearest integers (Ii) to the corrected residue number (kbstni) are those listed in column six of Table VI1 as minimum residue numbers and compared with those reported by Vandekerckhove and Weber (1978). This method yielded a total of 373 amino acid residues/mol of porcine G-actin and a molecular weight of 41 648 consistent with the values 41 871 and 375 amino acid residues reported by Vandekerckhove and Weber (1978). The results are also consistent with the currently favored concept that eukaryotic actins are highly conserved in evolution (Pollard and Cooper, 1986). Porcine G-actin from the semimembranosus muscle contained 0.97 mol of His(s-Me)/mol of protein (Table VII). Typical chromatographicseparations obtained, when

+

a synthetic mixture of all methylated basic amino acids and porcine actin from the semimembranosusmuscle were analyzed on the 0.28 X 50 cm column of Dionex DC-4A resin by the method described previously (Zarkadas et al., 1987b), are illustrated in Figure 3. As may be seen in Figure 3B, the analysis of a 96-h hydrolysate of porcine actin (100 pL equivalent to 150 pg protein per analysis) revealed the complete separation, at pH 4.501, of His(7Me) following the ammonia peak, along with five as yet unidentified ninhydrin positive peaks, designated 1-5. The two major unknown compounds, 3 and 5, present in relatively large amounts in porcine actin have not been previously reported, and further investigations will be necessary for their isolation and characterization. These results are in close agreement with those reported by Elzinga et al. (1973), Vandekerckhove and Weber (1978,1979),and Vandekerckhove et al. (1986), and for purposes of comparison, their data are included in Table VII. From these results it can be concluded that the lower protein-bound His(s-Me) content of the porcine semimembranosus muscle could be attributed to differences in the distribution of myosin heavy-chain isoforms in this skeletal muscle tissue. Staron and Pette (1987a,b) have shown that a maximum of 54-60 myosin isoforms can coexist in just two rabbit skeletal muscles (soleus and tibalis anterior). From this experimental evidence it became apparent that the composition of muscle tissues is much more complex than anticipated earlier, and further studies are required. The recent sequence studies of Maita et al. (1987), for example, have indicated that the His(s-Me) is located at position 755 in the heavy-polypeptide chains of myosin rather than near the active site of this protein, at position 69, as originally reported by Elzinga and Collins (1977) and Huszar (1984). From amino acid and sequence data, spectral measurements, and radioactive labeling, Okamoto and Yount (1985) have shown that the location of the unusual amino acid IP,IP,IP-trimethyllysinenext to Trp-130 suggest that the positively charge [Lys(Me,)+] residue may provide part of the binding site for the triphosphate portion of ATP.

1142 J. Agric. Food Chem., Vol. 36, No. 6, 1988

Zarkadas et ai.

Table IX. Myofibrillar and Connective Tissue Protein Contents of Selected Porcine Skeletal Muscles from Mature Yorkshire Sows % of total muscle proteins

skeletal muscle protein fraction"

ii

myofibrillar* actin mysoin actomyosin other sol proteins

iii iv v vi

connective tissueC collagend elastine total iv + v

i

psoas maior (rabbit) protein, g/kg total proteins external Yates and Hanson and external sternomansemitensemimemsternoman- semiten- semimem- Greaser Hyxley dibularis dinosus branosus dibularis dinosus branosus (1983) (1957) F1, Intracellular: SDS-Soluble 580.75 f 8.83 577.81 f 9.81 511.10 f 4.91 52.14 54.63 48.45 57.71 62.00 122.54 i 1.86 121.92 f 2.07 107.85 f 1.04 11.00 11.53 10.22 12.69 12.00 255.15 f 3.88 253.86 f 4.31 224.55 f 2.16 22.91 24.00 21.29 24.82 34.00 377.70 f 5.74 375.78 f 6.38 332.40 f 3.19 33.91 35.53 31.51 37.52 46.00 419.25 422.19 488.90 37.64 39.92 46.35 34.00 895.35 f 15.02 576.66 f 3.17 14.04 f 0.12 590.70

F2, Extracellular Matrix: SDS-Insoluble 948.50 f 17.59 860.00 f 13.49 9.15 620.97 f 3.80 546.91 f 4.43 5.89 18.16 f 0.12 12.11 i 0.06 0.143 639.13 559.02 6.033

5.17 3.38 0.099 3.480

4.47 2.84 0.063 2.903

"Separated by the combined procedures of McCollester (1962) and Laurent et al. (1981). bCalculated from eq 6a, 6b, 8, and 8a-c. cCalculated from Pro(4-OH) data (Table V) by eq 5c. dCalculated from Lys(5-OH) data (Table V) by eq 5a with the data taken from Miller and Gay (1982) and Light et al. (1985). eCalculated from eq 5b with data taken from Foster (1982).

Determination of the Actin and Myosin Components of the Selected Porcine Skeletal Muscles. From the His(T-Me) contents of the SDS-solubilized intracellular muscle protein fractions (Fl) isolated from three typical skeletal muscles of mature Yorkshire sows (Table IV), it has been possible to calculate the myosin, actin, actomyosin, and myofibrillar proteins present with eq 8-8c, and the results are summarized in Table IX. The data expressed as grams of protein per kilogram of total protein show deviations of less than 100 f 3.0% from the average values. In porcine skeletal muscles actin accounts for an estimated 11.00-11.53% of the total muscle protein or about 21.1 % of the myofibrillar proteins, and myosin accounts for another 21.3-24.0% of the total muscle mass corresponding to about 43.9% of the myofibrillar proteins (weighted mean 51.74%; range 48.45-54.63%). These values are in close agreement with those reported previously for the myofibrillar protein contents of the costal region of the bovine diaphragm and all-beef hamburger which were 52.26% and 52.31% of the total bovine muscle proteins, respectively (Zarkadas et al., 1988; Karatzas and Zarkadas, 1988). These results are also in accord with those reported by Yates and Greaser (1983) who have used different methods and different muscle tissue for these determinations. These authors have shown that the rabbit psoas muscle contains 57.71 % myofibrillar protein of the total muscle mass and that the myofibrils contain 22% actin and 43% myosin by weight. Hanson and Huxley (1957) found by quantitative extraction that the myofibril would contain 21% of actin and 43% of myosin weight, compared to the myosin value reported by Szent-Gyorgi et al. (1955) which accounts for 38% of the myofibrillar proteins. Although the relative amounts of actin and myosin per unit of tissue among the porcine muscles investigated remains constant, the small differences noted in myofibrillar protein content of semimembranosus from the average figure of 53.3% reported in Table IX may reflect variations in the levels of fast, slow, or mixed fast/slow classes of myosin heavy-chain isoforms present in this skeletal muscle. Recent evidence has indicated that all three adult forms of myosin in skeletal muscle, corresponding to two adult fast myosin heavy chains (fast oxidative IIA and fast glycolytic IIB) and one slow myosin heavy chain (slow oxidative, type I), are tissue-specific and developmentally regulated, with more than one myosin heavy-chain gene expressed in each muscle and developmental stage (Whalen, 1985; Lowey, 1986; Gauthier, 1986; Mahdavi et

al., 1986). It has also been shown that the slow (type I) myosin heavy chain is identical with the cardiac @-myosin heavy chain, since both isoforms are encoded by the same gene in slow-twitchskeletal muscles (soleus) of the mouse (Mahdavi et al., 1986). These findings indicate that the small differences noted in His(T-Me) content among the porcine skeletal muscles evaluated (Table VI) may reflect the amounts of slow-twitch or cardiac @-myosinheavy isoforms present in the semimembranosus muscle. This explanation is consistent with earlier studies by Huszar (1984) who has shown that His(T-Me) is absent from the myosin heavy chains of slow and cardiac muscles and in fast muscles of unborn and newborn animals. As mentioned earlier, further structural studies of myosin isoforms are required to clarify the early reports on the His(T-Me) contents of myosin from different muscles. Determination of the Connective Tissue Proteins. The extracellular matrix protein fraction (F2) of the porcine skeletal muscles, which is high in collagen and elastin content, was isolated and quantitated (Table 111) by the combined methods of McCollester (1962) and Laurent et al. (1981). Tables 111, V, VI, and VI11 show that the external sternomandibularis muscle contained the highest levels (10.2%) of extracellular matrix proteins, compared to the other two muscles, which ranged from 5.20 to 5.45 g of protein/100 g of total muscle protein. The contents of Des, iDes, Lys(bOH), and Pro(4-OH) per unit of the extracellular matrix protein (grams per kilogram of total protein), determined by the sensitive methods described previously (Zarkadas et al., l986,1987b), were used as markers for assessing the levels of the connective tissue proteins, i.e., collagen and elastin, in skeletal muscles. In this approach the content of collagen and collagen-like proteins (Porter and Reid, 1978; Reid, 1982; Anglister et al., 1976) can be calculated by multiplying the amounts of aLys(5-OH) found in the acid hydrolysates of fraction F2 by the conversion factor 63.3 (eq 5a) and the elastin content from the amounts of Des present (Gunja-Smith, 1985; Zarkadas et al., 1986, 1987b) by eq 5b. The total connective tissue protein content of the extracellular matrix has also been calculated from the amounts of Pro(4OH) found in tissue hydrolysates (Berg, 1982) by eq 5c. Table IX shows that the mean values for total collagen ranged from 2.84% in sow semimembranosus, 3.38% in semitendinosus, to 5.89% in the external sternomanidibularis muscle for sows. These results are in reasonably good agreement with those reported by Bendall (1967), Dransfield (1977), and Light et ai. (1985) for the distribution of collagen (average 4.35%; range 1.22-15.1%) in

Quantitation of Contractile Protein in Porcine Muscle

34 bovine skeletal muscles investigated. These results also demonstrate that the content of collagen is higher in slow-twitch postural muscles than that of porcine locomotory muscles, which contain primarily fast-twitch fiber types. This suggestion is supported by recent studies of Kovanen (1980, 1984) and others (Garcia-Bunnel and Garcie-Bunnel, 1967) who have shown that differences in muscle collagen content appear to be related to the level of individual muscle fibers or even at the level of perimysium and endomysium within a given fiber type. The highest level of elastin (0.143%) was found in the external sternomandibularis muscle, with the semitendinosus being much lower in total elastin (0.099%) compared to 0.063% found in the semimembranosus (Table IX). This datum was based on the Des value (3 residues/1000 total residues) reported by Foster (1982) for purified amorphous elastin. Bendall (1967) has shown that the semitendinosus bovine muscle, which contained exceptionally high levels of elastin (1.82%), had rather low levels of Des and iDes. If his results, as summarized in Table VIII, are used to quantitate elastin in the extracellular matrix of porcine muscles, the amounts of elastin would be about 40% higher. Rowe (1986) has provided evidence to show that both high (semitentinosus) and low (longissimus dorsi) elastin content bovine muscles appear to have the same anatomical organization in the two structural forms of elastin present in the epimysium and perimysium and that both elastin fibers are aligned with the long axis of the muscle fibers. The content of total connective tissue proteins in the semimembranosus, semitendinosus, and external sternomandibularis porcine muscles, estimated from the sums of collagen and elastin, were respectively 2.0, 3.48, and 4.03% (Table IX). The mean values for the transcellular matrix proteins that remained ranged from 1.97% in semitendinosus, with the external sternomandibularis being considerably higher in total transcellular proteins (4.19%), compared to 2.30% found in the semimembranosus. Loewy et al. (1983) reported that the transcellular matrix fiber ghost proteins in skeletal muscles account for 170of the original muscle proteins. These authors have shown that the transcellular matrix proteins are rich in glycine and alanine but very low in basic and acidic amino acids, they are covalently cross-linked with e-(7-glutamine)lysine, and they have little or no Pro(4-OH). These results suggest that, in addition to collagen and elastin, the extracellular matrix from porcine muscles contains various other proteins. Blumenfeld et al. (1983) indicated that the extracellular matrix in most tissues consists of collagen, elastin, microfibrillar proteins, proteoglycans (Carrino and Caplan, 1982,1986),and glycoproteins such as fibronectin or laminin, all of which are involved in maintaining proper alignment of the muscle fibers and in a variety of morphogenetic and developmental processes (Wiens et al., 1984). At each stage of development and maturation the extracellular matrix consists of an array of components whose specified compositions, amounts,and relative proportions are characteristic of the individual tissues. Little is known about how these macromolecules are assembled to form the matrix. On the basis of known distributions of Pro(4-OH) in elastin and the various collagen isotypes given in Table I and VIII, the contents of total connective tissue in porcine muscles could be calculated from the amounts of Pro(4OH). The results listed in Table IX indicate that the porcine muscles contained higher amounts of Pro(4-OH) than other vertebrate connective tissues and that this unique to collagen and elastin amino acid shows larger

J. Agric. Food Chem., Vol. 36, No. 6, 1988

1143

variation in level in porcine muscle than either Des or Lys(5-OH). Other Muscle-Soluble Proteins. From the foregoing results, it became apparent that, in addition to myofibrillar and connective tissue proteins of the extracellular matrix (Table IX), a large quantity of SDS-soluble proteins is found in the intracellular muscle protein fraction of porcine muscles which ranged from 37.64 to 46.4% of the total muscle proteins. These results are considerably higher than those found (2&34%) by Hanson and Huxley (1957) as soluble proteins (sacroplasmic) washed out of the glycerol-extracted muscle after it has been broken up into fibrils, but they are in accord with the 41.6% figure quoted by Szent-Gyorgi et al. (1955) using two different extraction procedures. These results suggest that the intracellular muscle protein fraction includes sarcoplasmic proteins, organelles, Z-band, and other membrane proteins, etc., which are soluble in 2% sodium dodecyl sulfate by the combined procedures of McCollester (1962) and Laurent et al. (1981). The results presented show the variation that exists in amino acid content among selected porcine muscles from mature Yorkshire sows that are being marketed today for meat processing. From these results, it is evident that the methods of analysis developed for determining all methylated basic amino acids, Lys(5-OH), the stable elastin cross-links Des and iDes, and related compounds in proteins and muscle tissues, can be easily applied for the determination of the myofibrillar and connective tissue protein contents of skeletal muscle tissues from porcine and the other major meat-yielding species. From these results it can also be concluded that a potentially useful means for evaluating the protein quality of different muscles or meats might be based on knowledge of their amino acid composition, as recommended by the Expert Work Group (FSIS, 1984) and Young and Pellett (1984). ACKNOWLEDGMENT

We are indebted to Professor P. H. Schuepp, the former Chairman of the Department of Agricultural Chemistry and Physics, Macdonald College of McGill University, where part of this research was conducted; Dr. L. P. Lefkovitch for the statistical design that made the testing of the concepts described in this paper possible; J. A. Rochemont for his expert advice on instrumentation; A. Lutz of the Art and Design Unit, Research Program Service Section (Agriculture Canada), for the illustrations; Dianne Martin Beyeler for typing this manuscript; and Danny Smith, Meat Inspector (Agriculture Canada) at Abattoir Regional, Valleyfield Inc., Valleyfield, Quebec, for providing the porcine skeletal muscles used in these studies. Registry No. His(T-Me), 332-80-9; Lys(5-OH), 1190-94-9; Pro(4-OH), 51-35-4;Lys(Me2), 2259-86-1; Lys(Me3), 23284-33-5; desmosine, 11003-57-9; isodesmosine, 991-01-5. LITERATURE CITED Anglister, L.; Rogozinski, S.; Silman, I. “Detection of 4Hydroxyproline in Preparations of Acetylcholinesterase from the Electric Organ of the Electric Eel”. FEBS Lett. 1976,69, 129- 132. Asatoor, A. M.; Armstrong, M. D. ““-Methylhistidine, a Component of Actin”. Biochem. Biophys. Res. Commun. 1967,26, 168-174. Ashworth, R. B. “Amino Acid Analysis for Meat Protein Evaluation“. J . Assoc. Off. Anal. Chem. 1987, 70, 80-85. Beecher, G. R.; Kastenschmidt, L. L.; Cassens, R. G.; Hoekstra, W. G.; Briskey, E. J. “A Comparison of the Light and Dark Portions of a Striated Muscle”. J. Food Sei. 1968, 33,84-88.

1144

J. Agric. Food Chem., Vol. 36,No. 6, 1988

Beerman, D. G.; Cassens, R. G.; Hausman, G. J. “A Second Look at Fiber Type Differentiation in Porcine Skeletal Muscle”. J. Anim. Sei. 1978,46, 125-132. Bendall, J. R. “The Elastin Content of Various Muscles of Beef Animals”. J. Sci. Food Agric. 1967, 18, 553-558. Benedict, R. C. “Determination of Nitrogen and Protein Content of Meat and Meat Products”. J. Assoc. Off. Anal. Chem. 1987, 70, 69-74. Berg, R. A. Methods Enzymol. 1982,82, 372-398. Black, L. W.; Hogness, D. W. “The Lysozyme of Bacteriophage h 11. Amino Acid and End Group Analysis”. J. Biol. Chem. 1969, 244, 1976-1981. Bligh, E. G.; Dyer, W. J. “A Rapid Method for Total Lipid Extraction and Purification”. Can. J . Biochem. Physiol. 1959, 37, 911-917. Block, R. J.; Mitchell, H. H. “The Correlation of the Amino Acid Composition of Proteins with their Nutritive Value”. Nutr. Abstr. Rev. 1946, 16, 249-278. Blumenfeld, D. 0.; Bienkowski, R. S.; Schwartz, E.; Seifter, S. Extracellular Matrix of Smooth Muscle Cells; Stephens, N. L., Ed.; CRC Biochemistry of Smooth Muscle 11;CRC: Boca Raton, FL, 1983; pp 137-187. Carnegie, P. R.; Hee, K. P.; Bell, A. W. “Ophidine alanyl-^“-Methylhistidine, ‘Balenine’)and other Histidine Dipeptides in Pig Muscles and Tinned Hams”. J . Sci. Food Agric. 1982, 33, 795-801. Carnegie, P. R.; Collins, M. C.; Ilic, M. Z. “Use of Histidine Dipeptides to Estimate the Proportion of Pig Meat in Processed Meats”. Meta Sci. 1984, 10, 145-154. Carrino, D. A.; Caplan, A. I. ”Isolation and Preliminary Characterization of Proteoglycans Synthesized by Skeletal Muscle”. J. Biol. Chem. 1982,257, 14145-14154. Carrino, D. A.; Caplan, A. I. Proteoglycan Synthesis During Skeletal Muscle Development; Emerson, C., Fischman, D., Nadal-Ginald, B., Siddiqui, M. A. Q., Eds.; Molecular Biology of Muscle Development; Alan R. Liss: New York, 1986; pp 117-131. Dransfield, E. J. “Intramuscular Composition and Texture of Beef Muscles”. J . Sci. Food Agric. 1977, 28, 833-842. Eastoe, J. E. “Composition of Collagen and Allied Proteins”. In Treatise on Collagen; Ramachandran, G. N., Ed.; Academic: New York, 1967; pp 1-72. Elzinga, M.; Collins, J. H. “Amino Acid Sequence of a Myosin Fragment that contains SH1, SH2 and Nr-Methylhistidine”. Proc. Natl. Acad. Sci. U.S.A. 1977, 74, 4281-4284. Elzinga, M.; Collins, J. H.; Kuehl, W. M.; Adelstein, R. S. ”Complete Amino Acid Sequence of Actin of Rabbit Skeletal Muscle”. Proc. Natl. Acad. Sci. U.S.A. 1973, 70, 2687-2691. Etherington, D. J.; Sims, T. J. “Detection and Estimation of Collagen”. J . Sci. Food Agric. 1981, 32, 539-546. “Final Report and Expert Work Group (FSIS). Recommendations”. Amer. J . Clin. Nutr. 1984,40,675-684. Foster, J. S. Methods Enzymol. 1982, 82, 559-570. Gardia-Bunuel, L.; Garcia-Bunuel, V. M. “Connective Tissue and the Pentose Phosphate Pathway in Normal and Denervated Muscle”. Nature (London) 1967,213, 913-914. Gauthier, G. F. Skeletal Muscle Fiber Types; Engel, A. G., Banker, B. Q., eds.; Myology I; McGraw-Hill: New York, 1986; pp 255-283. Gauthier, G. F.; Lowey, S. “Distribution of Myosin Isoenzymes Among Skeletal Muscle Fiber Types”. J . Cell. BioE. 1979, 81, 10-25. Griffith, D. W. P-Amino Acids; Mammalian Metabolism and Utility as a-Amino Acid Analogues; Richardson, C. C., Boyer, P. D., David, I. R., Meister, A,, Eds.; Annual Review of Biochemistry 55; Annual Reviews: Palo Alto,CA, 1986; pp 855-878. Gunja-Smith, Z. “An Enzyme-linked Immunosorbent Assay to Quantitate the Elastin Crosslink Desmosine”. Anal. Biochem. 1985, 147, 258-268. Hancock, W. S.; Harding, D. R. K. Review of Separation Conditions; Hancock, W. s.,Ed.; CRC Handbook of HPLC for the Separation of Amino Acids, Peptides, and Proteins I; CRC: Boca Raton, FL, 1986; pp 235-262. Hanson, J.; Huxley, H. E. “Quantitative Studies on the Structure of Cross-striated Myofibrils 11. Investigations by Biochemical Techniques”. Biochim. Biophys. Acta 1957, 23, 250-260.

Zarkadas et al. Happich, M. L.; Ackerman, S.; Miller, A. J.; Swift, C. E.; Gumbman, M. R. “Composition and Protein Efficiency Ratio of Meat Samples Partially Defatted with Petroleum Ether, Acetone, or Ethyl Ether”. J . Assoc. Off. Anal. Chem. 1984, 67, 265-270. Harris, C. I.; Milne, G. “The Urinary Excretion of “-Methylhistidine by Cattle: Validation as an Index of Muscle Protein Breakdown”. Br. J. Nutr. 1981,45, 411-422. Harris, C. I.; Milne, G. “The Identification of the “-Methylhistidine-containing Dipeptide, Balenine, in Muscle Extracts from Various Mammals and the Chicken”. Comp. Biochem. Physiol. 1987, 86B, 273-279. Haverberg, L. N.; Onstedt, P. T.; Munro, H. N.; Young, V. R. “W-Methylhistidine Content of Mixed Proteins in Various Rat Tissues”. Biochim. Biophys. Acta 1975, 405, 67-71. Heidelbaugh, N. D.; Huber, C. S.; Bernarczyk, J. R.; Smith, M. D.; Rambant, P. C.; Sheeler, H. 0. “Comparison of Three Methods for Calculating Protein Content of Foods”. J. Agric.

Food Chem. 1975,23,611-613. Hibbert, I.; Lawrie, R. A. ”Technical Note: The Identification of Meat in Food Products”. J.Food Technol. 1972,23,333-335. Holbrook, I. B.; Gross, E.; Irving, M. H. “NT-Methylhistidinein Human Skeletal and Smooth Muscle Proteins”. Br. J . Nutr. 1979,41, 15-17. Horstmann, J. H. “A Precise Method for the Quantitation of Proteins Taking into Account their Amino Acid Composition”. Anal. Biochem. 1979, 96, 130-138. Hoy, T. G.; Ferdinand, W.; Harrison, P. M. “A Computer-assisted Method for Determining the Nearest Integer Ratios of Amino Acid Residues in Purified Proteins”. Int. J. Peptide Protein Res. 1974, 6 , 121-140. Hugli, T. E.; Moore, S. “Determination of Tryptophan Content of Proteins by Ion-Exchange Chromatography of Alkaline Hydrolysates”. J. Biol. Chem. 1972, 247, 2828-2834. Hunt, S. “Degradation of Amino Acids Accompanying in vitro”. In Chemistry and Biochemistry of Amino Acids; Barrett, G. C., Ed.; Chapman and Hall London, 1985; pp 376-414. Huszar, G. Methods Enzymol. 1984, 106, 287-295. Johnson, L. S. “Nr-Methylhistidine in Actin and Other Muscle Proteins”. Biochin. Biophys. Acta 1967, 105, 361-370. Jones, A. D.; Homan, A. C.; Favell, D. J.; Hitchcock, C. H. S. “Investigation of the Levels of W-Methylhistidine in a Range of Beef Cuts and Offals”. Meat Sci. 1985, 15, 137-147. Jones, A. D.; Homan, A. C.; Favell, D. J. “Investigation of the Levels of W-Methylhistidine in a Range of Pork and Chicken Meats”. Meat Sei. 1987, 19, 197-205. Karatzas, C. N.; Zarkadas, C. G. “The Determination of the myofibrillar and Connective Tissue Protein Contents and Amino Acid Composition of Selected Composite Meat Products”. J. Agric. Food Chem. 1988, companion paper in this issue. Kiessling, K.-H.; Hanson, I. “Fibre Composition and Enzyme Activities in Pig Muscles”. Swedish J . Agric. Res. 1983, 13, 257-261. Kiessling, K.-H.; Lundstrom, K.; Petersson, H.; Stalhammer, H. “Age and Feed-related Changes of Fibre Composition in Pig Muscle”. Swedish J. Agric. Res. 1981, 12, 69-75. Kovanen, V.; Suominen, H.; Keikkinen, E. ”Connective Tissue of Fast and Slow Skeletal Muscle in Rats Effects of Endurance Training”. Acta Physiol. Scand. 1980, 108, 173-180. Kovanen, V.; Suominen, H.; Heikkinen, E. “Collagen of Slow and Fast Twitch Muscle Fibres in Different Types of Rat Skeletal Muscle”. Eur. J . Appl. Physiol. 1984, 52, 235-242. Laemmli, U. K. “Cleavage of Structural Proteins During the Assembly of the Head of Bacteriophage T4”. Nature (London) 1970,227, 680-685. Laurent, G. J.; Cockerill, P.; McAnulty, R. J.; Hastings, J. R. B. “A Simplified Method for the Quantitation of the Relative Amounts of Type I and Type I11 Collagen in Small Tissue Samples”. Anal. Biochem. 1981,113, 301-312. Light, N. D. “Collagen in Skin: Preparation and Analysis”. In Methods in Skin Research; Skerow, D., Skerow, C. J., Eds.; Wiley: New York, 1985. Light, N.; Champion, A. E. “Characterization of Muscle Epimysium, Perimysium and Endomysium Collagens”. Biochem. J . 1984,219, 1017-1026.

Quantitation of Contractile Protein in Porcine Muscle Light, N.; Champion, A. E.; Voyle, C.; Bailey, A. J. “The Role of Epimysial, Perimysial and Endomysial Collagen in Determining Texture in Six Bovine Muscles”. Meat Sei. 1985,13,137-149. Loewy, A. G.; Wilson, F. J.; Taggart, F. J.; Greene, E. A.; Frasca, P.; Kaufman, H. S.; Sorrell, M. J. “A Covalently Cross-linked Matrix in Skeletal Muscle Fibers”. Cell Motility 1983, 3,

463-483. Lowey, S. “Cardiac and Skeletal Muscle Myosin Polymorphism”. Med. Sci. Sports Exercise 1986a,18, 284-291. Lowey, S. The Structure of Vertebrate Muscle Myosin; Engel, A. G.; Banker, B. Q., Eds.; Myology I; McGraw-Hilk New York, 1986b;pp 563-587. Lowry, 0. H.; Rosebrough, N. J.; Farr, A.; Randall, R. J. ”Protein Measurement with the Folin Phenol Reagent”. J. Biol. Chem.

1951,193,265-275. Mahdavi, V.; Strehler, E. E.; Periasamy, M.; Wreczarek, D.F.; Izumo, S.; Nadal-Ginard, B. “Sarcomeric Myosin Heavy Chain Gene Family: Organization and Pattern of Expression”. Med. Sei. Sports Exercise 1986, 18, 299-308. Maita, T.; Hayashida, M.; Tanioka, Y.; Konine, Y.; Matsuda, G. “The Primary Structure of the Myosin Head”. Proc. Natl. Acad. Sci. U.S.A. 1987,84,416-420. McCollester, D. L. “A Method for Isolating Skeletal Muscle Cell Membrane Components”. Biochim. Biophys. Acta 1962,57,

427-437. McCollester, D. L.; Semente, G. “Membrane Isolation and Cytoskeletal Breakdown. I. The Effect of Endogenous Enzymes, Ions, pH, Temperature and Dissolved Gases”. Biochim. Biophys. Acta 1964,90,146-158. McNeal, J. E. “Rapid Methods for Determination of Meat Composition“. J. Assoc. Off. Anal. Chem. 1987, 70, 95-99. Miller, E. J.; Gay, S. Methods Enzymol. 1982,82, 3-32. Moore, S. “On the Determination of Cysteine and Cysteic Acid”. J . Biol. Chem. 1963,238,235-237. Murakami, U.; Uchida, M. “Contents of Myofibrillar Proteins in Cardiac, Skeletal, and Smooth Muscle”. J . Biochem. 1985,98,

187-197. Nguyen, Q.; Fanous, M. A.; Kamm, L. H.; Khalili, A. D.; Schuepp, P. H.; Zarkadas, C. G. “Comparison of the Amino Acid Composition of Two Commercial Porcine Skins (Rind)”. J. Agric. Food Chem. 1986,34,565-572. O’Farrell, P. H. “High Resolution Two-Dimensional Electrophoresis of Proteins”. J. Biol. Chem. 1975,250,4007-4021. Okamoto, Y.; Yount, R. G. “Identification of an Active Site Peptide of Skeletal Myosin after Photoaffinity Labeling with N-(4azido-2-nitrophenyl)-2-aminoethyl diphosphate”. Proc. Natl. Acad. Sei. U.S.A. 1985,82,1575-1579. Olsman, W. J.; Slump, P. Methods of Determination of Connective Tissue Free Muscle Protein in Meat Products; Lawrie, R., Ed.; Developments in Meat Science 2; Applied Science: NJ, 1981;pp 197-240. Oser, B. L. ”Methods for Integrating Essential Amino Acid Content in the Nutritional Evaluation of Protein”. J. Am. Diet. ASSOC.1951,27,396-402. Pardee, J. D.; Spudich, J. A. Purification of Actin; Frederiksen, D. W., Cunningham, L. W., Eds.; Academic: New York, 1982; pp 164-181. Peterson, G. L. Methods Enzymol. 1983,91,95-119. Pollard, T. D.; Cooper, J. A. Actin and Actin-binding Proteins. A Critical Evaluation of Mechanisms and Functions; Richardson, D. D., Boyers, P. D., David, C. B., Meister, A., Eds.; Annual Review of Biochemistry 55;Annual Review: Palo Alto, CA, 1986;pp 987-1035. Porter, P. R.; Reid, K. B. M. “The Biochemistry of Complement”. Nature (London) 1978,275,699-704. Rangeley, W.R.D.; Lawrie, R. A. “Methylamino Acids as Indices in Meat Products. I. The Development and Validity of an Analytical Procedure”. J. Food Technol. 1976,11,143-159. Rangeley, W. R. D.; Lawrie, R. A. “Methylamino Acids ag Indices in Meat Products. 11. Further Examination of Protein Sources and the Practical Application of Methylamino Acid Titres in Predicting Meat Content”. J. Food Technol. 1977,12,9-26. Ranken, M. D. Composition of Meat: Some Structural and Analytical Implications; Hudson, B. J. F., Ed.; Developments in Food Proteins 3; Elsevier Applied Science: New York, 1984; pp 1-32.

J. Agric. Food Chem., Vol. 36,No. 6, 1988

1145

Reid, K. B. M. Methods Enzymol. 1982,82,319-324. Rowe, R. W.D. “Elastin in Bovine Semitendinosus and Longissimus dorsi Muscles”. Meat Sci. 1986,17, 293-317. Rucker, R. B. Methods Enzymol. 1982,82,650-657. Spudich, J. A.; Watt, S. “The Regulation of Rabbit Skeletal Muscle Contraction. I. Biochemical Studies of the Interaction of the Tropomyosin-Troponin Complex with Actin and the Proteolytic Fragments of Myosin”. J. Biol. Chem. 1971,246,4866-4871. Staron, R. S.;Pette, D. “The multiplicity of Combinations of Myosin Light Chains and Heavy Chains in Histochemically Typed Single Fibers. Rabbit Soleus Muscle”. Biochem J.

1987a,243,687-693. Staron, R. S.;Pette, D. “The Multiplicity of Combinations of Myosin Light Chains and Heavy Chains in Hystochemically Typed Single Fibres. Rabbit Tibalis anterior muscle”. Biochem. J. 1987b,243,695-699. Statistical Analysis System (SAS) User’s Guide: Basics; SAS Institute: Cary, NC, 1982. Suzuki, J. A.; Cassens, R. G. “A Histochemical Study of Myofiber Types in Muscle of the Growing Pig”. J . Anim. Sci. 1980,51,

1449-1461. Szent-Gyorgi, A. G.;Mazia, D.; Szent-Gyorgi, A. “On the Nature of the Cross-striation of Body Muscle”. Biochim. Biophys. Acta

1955,16,339-342. Tong, S. W.; Elzinga, M. “The Sequence of the NH2-Terminal 204-residue Fragment of the Heavy Chain of Rabbit Skeletal Muscle Myosin”. J. Biol. Chem. 1983,258,13100-13110. Tristram, G. R.; Smith, R. H. The Amino Acid Composition of Some Purified Proteins; Anfinsen, C. B., Anson, M. L., Edsall, J. T., Eds.; Advances in Protein Chemistry 18;Academic: New York, 1963;pp 227-318. Vandekerckhove, J.; Weber, K. “At Least Six Different Actins are Expressed in a Higher Mammal: An Analysis Based on the Amino Acid Sequence of the Amino Terminal Tryptic Peptide”. J. Mol. Biol. 1978, 126, 783-802. Vandekerckhove, J.; Weber, K. “The Complete Amino Acid Sequence of Actins from Bovine Aorta, Bovine Heart, Bovine Fast Skeletal Muscle, and Rabbit Slow Skeletal Muscle”. Differentiation 1979,14, 123-133. Vandekerckhove,J.; Bugaisky, G.; Buckingham, M. “Simultaneous Expression of Skeletal Muscle and Heart Actin Proteins in Various Striated Muscle Tissues and Cells”. J. Biol. Chem.

1986,261, 1838-1843. Weber, K.; Osborn, M. Proteins and Sodium Dodecyl Sulfate: Molecular Weight Determinution on Polyacrylamide Gels and Related Procedures; Neurath, H. G., Hill, R. L., Boeder, C. L., Eds.; The Proteins 1;Academic: New York, 1975;pp 179-223. Whalen, R. G. “Myosin Isoenzymes as Molecular Markers for Muscle Physiology”. J.Exp. Biol. 1985,115, 43-53. White, W.J. P.; Lawrie, R. A. “Variations in the Levels of W Methylhistidine of the Myosins within the Bovine Carcass”. Meat Sei. 1985,15, 173-181. Wiens, D.; Sullins, M.; Spooner, B. S. “Precardiac Mesoderm Differentiation in Vitro: Actin-Isotype Synthetic Transitions, Myofibrillogenesis, Initiation of Heartbeat and the Possible Involvement of Collagen”. Differentiation 1984,28, 67-72. Yates, L. D.; Greaser, M. L. “Quantitative Determination of Myosin and Actin in Rabbit Skeletal Muscle and Myofibrils”. J. Mol. Biol. 1983,168, 123-141. Young, V. R.; Pellett, P. L. “Background Paper 5: Amino Acid Composition in Relation to protein Nutritional Quality of Meat and Poultry Products”. Am. J. Clin. Nutr. 1984,40,737-742. Zarkadas, C. G. “A Simple Chromatographic Method for the Determination of the Methylated Basic Amino Acids in Proteins”. Can J . Biochem. 1975,53,96-101. Zarkadas, C. G. “Accelerated Methods for Determining Methylated Basic Amino Acids in Proteins and Muscle Tissues and Stable Crosslinks in Elastin and Collagen”. J . Appl. Biochem.

1979,1, 148-163. Zarkadas, C. G. “The Determination of the Myofibrillar and Connective Tissue Contents of Meats and Composite Meat Products”. Proceedings of the 27th European Fleischforscher-Kongress 11, Vienna, Austria, 1981;pp 572-575. Zarkadas, C. G.;Zarkadas, G. C.; Karatzas, C. N.; Khalili, A. D.; Nguyen, Q. “Rapid Method for Determining Desmosine, Isodesmosine, 5-Hydroxylysine, Tryptophan, Lysinoalanine and

1146

J. Agric. Food Chem. 1900, 36, 1146-1 150

the Amino-sugars in Proteins and Tissues". J . Chromatogr. 1986,378,67-76. Zarkadas, C. G.; Marshall, W. D.; Khalili, A. D.; Nguyen, Q.; Zarkadas, G. C.; Karatzas, C. H.; Khanizadeh, S. "Mineral Composition of Selected Bovine, Porcine and Avian Muscles, and Meat Products". J . Food Sci. 1987a, 52, 520-525. Zarkadas, C. G.; Rochemont, J. A.; Zarkadas, G. C.; Karatzas, C. N.; Khalili, A. D. "Determination of Methylated Basic, 5-

Hydroxylysine,Elastin Crosslinks, Other Amino Acids and the Amino Sugars in Proteins and Tissues". Anal. Biochem. 1987b, 160, 251-266.

Zarkadas, C. G.; Meighen, E. A.; Zarkadas, G. C.; Karatzas, C. N.; Rochemont, J. A.; Berthelet, M. "The Determinationof the Myofibrillar and Connective Tissue Proteins in the Bovine Diaphragm". J. Agric. Food Chem. 1988, companion paper in this issue. Received for review May 11,1987. Accepted January 26,1988. This investigation was supported by Agriculture Canada in the form of a DSS Contract (No. OSU81-00523)with McGill University and by McGill University in the form of postgraduate bursary awarded t o C.N.K. (1984-1985).

Levels of Formaldehyde in Milk, Blood, and Tissues of Dairy Cows and Calves Consuming Formalin-Treated Whey' Katherine E. Buckley,* Lorne J. Fisher, and Vernon G . MacKay2 Residual formaldehyde (CHzO) in the tissues, blood, and milk of calves and cows fed whey preserved with various levels of formalin was measured by gas chromatography with an electron-capture detection (ECD) system after derivatization with (2,4-dinitrophenyl)hydrazine.The morning milk from cows fed whey preserved with 0.05,0.10, and 0.15% formalin in three individual trials contained an average of 0.034, 0.095, and 0.208 mg of CHzO kg-', respectively. Endogenous levels of CH20 in milk from cows fed whey without added formalin were below the detection limit of 0.026 mg of CHzO kg-'. Average blood CHzO concentration in cows fed 0.15% formalin-treated whey was greater ( P < 0.01) than that of control cows a t 33 days (0.831 f 0.132 mgkg-' vs. 0.615 f 0.110 mgkg-') but not significantly ( P > 0.05) different 9 days (0.825 f 0.061 mg-kg-l vs 0.766 f 0.045 mgkg-') after formalin treatment was initiated. Fresh muscle tissue from calves consuming whey containing 0.10% formalin had a significantly ( P < 0.05) higher concentration of CH20 (0.256 mg-kg-') than fresh muscle tissue from control calves or calves consuming whey containing 0.05% formalin (0.178 and 0.206 mgkg-', respectively). There was no significant ( P > 0.05) effect due to formalin preservation of whey on CHzO concentrations in blood and frozen muscle, kidney, liver, and heart tissues. Production of liquid whey from the manufacture of cheddar cheese (sweet whey) and cottage cheese (acid whey) exceeds the demand for whey products to be incorporated into animal and human foods. Without costly treatment prior to disposal, this excess energy-rich resource becomes a serious environmental pollutant. Ideally, all liquid whey could be fed directly to livestock, with benefits accruing to the dairy industry by eliminating current disposal costs. Unfortunately centers of livestock production are often not in close enough proximity to cheese plants to allow daily delivery of fresh whey at an attractive cost to producers. Whey is an unstable feedstuff and should be offered fresh to maintain palatability. An alternative to daily whey delivery would be to add a preservative to the whey and store it in covered tanks. The addition of formalin (37% formaldehyde [CH20], w/w) a t a level of 0.1% by weight inhibits lactose conversion in whey with no reported detrimental effects on health and growth of livestock (Modler et al., 1980). The metabolic fate of ingested CH20 in livestock species is largely unknown, as is the effect of ingested CH20 on concentration of CH20 in vivo and the potential health risks to humans from chronic exposure to food products containing CH20. Interest by the dairy industry in obtaining government approval of formalin as a preservative Agriculture Canada, P.O. Box 1000, Agassiz, British Columbia, Canada VOM 1AO. Contribution No. 365. Present address: 46088 Greenwood Dr., Sardis, BC, Canada V2R 124. 0021-8561/88/1436-1146$01.50/0

for whey led to support for an investigation of the transfer of CHzO from ingested formalin-treated whey to tissues and milk of dairy cattle. The purpose of this investigation was to apply a new analytical technique to the determination of free and reversibly bound CH20 in tissues and milk of dairy animals ingesting formalin-treated acid whey under conditions simulating a commercial operation. Although the chromotropic acid assay has been a popular method for analysis of CH20 in livestock tissues and milk (Mills et al., 1972; Kreula and Rauramaa, 1976; Florence and Milner, 1981; Syrjala-qvist and Setiila, 1982a,b), the sensitivity of this method has been considered inadequate for determination of endogenous CH20 in animal tissues (Heck et al., 1982). Analysis of aldehydes as derivatives of (2,4-dinitropheny1)hydrazine by gas chromatography with 63Nielectron-capture detection [ECD] (Buckley et al., 1986) provides the sensitivity and specificity required to detect endogenous levels of CH20. EXPERIMENTAL SECTION Animals and Diets. Lactating Cows. The experiment involved the use of 12 Holstein dairy cows in their first trimester of lactation with a history of zero to eight previous lactations (group average of three lactations) and no previous exposure to CH20-treatedfeeds. The cows were trained to electronic gates (Calan Broadbent Doors, American Calan, Inc., Northwood, NH) to ensure that each animal received the designated treatment. The diet consisted of a low-enery pelleted concentrate (Table I) fed a t a level of 1kg of concentrate/3 kg of milk, liquid acid whey (Table 11),and good-quality grass hay offered ad libitum.

Published 1988 by the

American Chemical Society